The Odyssey of: Shape Matching with Ordered Boundary Point Shape Contexts Using a Least Cost Diagonal Method By Dr. Carl E Abrams March 14th, 2015 1 •Odyssey: a long wandering and eventful journey Or If we knew what we were doing, it wouldn't be called research, would it? – A Einstein 2 Agenda • Topic Selection • The Eventful Journey – Elation – Mild Desperation – Elation – Desperation (and beyond) – Relief (success) • Discussion / Lessons Learned 3 Topic Selection or… • First Idea: Requirements Engineering and Business Process Modeling • I despised the topic area 4 Topic Selection or…Roots • Identification of Automotive Vehicles Using Semantic Feature Extraction (Dec 2004) Elation! 5 Dissertation Timeline Carl E. Abrams Draft Chap 4&5 Now! Draft Chap 3&4 Defense 1/06 1/05 1/04 9/03 Draft Idea Paper Complete Draft Proposal Advisor Selection Complete Dissertation First Paper at Proposal Pace Day Final Manuscript and Paper Committee Formation Final Draft Chap 1-3 Draft Chap 1&2 6 The Eventful Journey 7 Semantic Geometric Features: A Preliminary Investigation of Automobile Identification Carl E. Abrams Sung-Hyuk Cha, Michael Gargano, and Charles Tappert 8 Agenda • Overview of the Problem • The Experiments • Results • Going Forward 9 Overview • Object recognition remains a hard problem • The human mind uses shapes to recognize objects • Can semantic features defined by their shapes be more effective in the recognition and identification of objects? 10 The Experiments • 10 test images of cars • Directly form the manufactures websites • Images were restricted to side views of the cars taken from 90 degrees • All 2005 models • Feature vectors calculated/measured from the images 11 The Vehicles Honda Accord Sedan 2005 Honda Civic Coupe 2005 Mazda 3 2005 Mazda 6 2005 Porsche Carrera Toyota Camry 2005 Toyota Corolla CE 2005 Toyota Celica GT 2005 Toyota Echo 2005 VW Passat 2005 12 Experiments used Euclidean Distance as the Measure d L (x t ) i 1 i 2 i the xi and ti are measurements from two different vehicles 13 Experiments used Euclidean Distance as the Measure (x2,y2) c b (x1,y1) a c = (a2+b2)1/2 c = ((x1-x2)2+(y1-y2)2)1/2 14 Manufacturers Specifications First Experiment Vehicle Description Wheelbase in inchesWidth Honda Accord Civic Coupe 2005 103.1 66.7 Honda Accord Sedan 2005 107.9 71.5 Mazda 3 2005 103.9 69.9 Mazda 6 2005 105.3 70.1 Porsche Carerra 2005 92.52 71.18 Toyota Camry 2005 107.1 70.7 Toyota Celica GT 2005 102.4 68.3 Toyota Corolla CE 2005 102.4 66.9 Toyota Echo 2005 92.3 65.4 VW Passat 2005 106.4 68.7 Length 175.4 189.5 178.3 186.8 174.29 189.2 170.5 178.3 164.8 185.2 Height 55.1 57.1 57.7 56.7 51.57 58.7 51.4 58.5 59.4 57.6 15 Boundary Description using Rays Second Experiment Vehicle Description Honda Accord Civic Coupe 2005 Honda Accord Sedan 2005 Mazda 3 2005 Mazda 6 2005 Mercedes Benz C230 Sports Coupe 2005 Nissan Altima 2005 Porsche Carerra 2005 Toyota Camry 2005 Toyota Celica GT 2005 Toyota Corolla CE 2005 Toyota Echo 2005 VW Passat 2005 Wheelbase in mm 148 139 146 140 138 136 128 136 134 144 154 145 Ray length in mm at angles in degress 5 10 15 20 125 121 123 110 120 116 118 110 118 113 115 113 118 104 115 109 102 115 112 115 115 110 110 110 112 112 115 100 126 127 126 120 102 104 111 113 119 117 120 113 125 124 127 112 122 120 120 110 25 95 98 100 95 100 91 90 101 87 96 113 96 30 90 90 94 88 94 86 82 94 82 93 108 90 35 86 85 89 84 89 81 76 88 75 89 100 86 40 81 80 84 80 82 76 71 86 68 84 99 82 45 78 78 80 76 79 71 67 73 65 82 96 74 50 75 73 78 72 73 66 66 73 63 77 92 72 55 70 70 71 69 71 63 62 71 60 73 87 69 60 66 67 67 63 67 62 60 68 57 69 84 66 16 Semantic Features Third Experiment Vehicle Description Honda Accord Civic Coupe 2005 Honda Accord Sedan 2005 Mazda 3 2005 Mazda 6 2005 Mercedes Benz C230 Sports Coupe 2005 Wheelbase in mm Angle FD Length FD Angle RD Length RD Angle Mirror Length Mirror 148 123 33 0 0 51 50 139 146 140 98 92 93 28 30 29 154 150 151 63 58 64 51 51 47 47 50 48 138 126 37 0 0 52 43 Nissan Altima 2005 Porsche Carerra 2005 136 98 25 153 58 47 45 128 144 33 0 0 69 34 Toyota Camry 2005 Toyota Celica GT 2005 Toyota Corolla CE 2005 Toyota Echo 2005 VW Passat 2005 136 85 24 151 62 42 53 134 125 31 0 0 49 44 144 154 145 90 97 97 34 34 36 148 0 154 63 0 59 42 49 47 58 65 50 17 Challenge: Determine the qualitative ability of the feature vectors to separate the vehicles • Within each experiment compute the distance of each vehicle from all the others • Evenly divide the measures into 5 bins • Observe the distribution of the measures 18 The Results 50 45 40 35 30 25 20 15 10 5 0 HWD Semantic Rays Bin1 Bin2 Bin3 Bin4 Bin5 19 Distance Matrix – Semantic Features Honda Civic Honda Accord Mazda 3 Mazda 6 Porsche Carerra Toyota Camry Toyota Celica Toyota Corolla Toyota Echo VW Passat Honda Civic 0 156.02 153.17 154.00 27.66 155.96 2.82 151.90 26.07 156.23 Honda Accord 156.02 0 7.21 7.07 161.72 16.09 156.36 13.45 154.01 4.12 Mazda 3 153.17 7.21 0 4.24 159.78 11.44 153.60 9.43 150.09 7.55 Mazda 6 154.00 7.07 4.24 0 160.89 9.43 154.37 6.55 151.06 5.00 Porsche Carerra 27.65 161.73 159.77 160.89 0 164/35 27.58 159.84 51.08 162.50 Toyota Camry 155.96 16.09 11.44 9.43 164.35 0 156.36 5.83 151.63 13.34 Toyota Celica 2.82 156.36 153.60 154.36 27.58 156.36 0 152.24 28.00 156.53 Toyota Corolla 151.90 13,45 9.43 6.55 159.84 5.83 152.24 0 148.33 10.48 Toyota Echo 26.07 154.01 150.09 151.06 51.07 151.63 28.00 148.33 0 154.01 VW Passat 156.23 4.12 7.55 5.00 162.50 13.34 156.53 10.48 154.01 0 20 Question #1 for the Class • What is the difference between the human hand and an automobile? • Answer: Nothing! • Commit the following to memory: A novel application of existing methods does not a dissertation topic make 21 End of First Dead End Mild Desperation 22 Dissertation Timeline Carl E. Abrams Mild Desperation sets in Draft Chap 4&5 Now! Draft Chap 3&4 Defense 1/06 1/05 1/04 9/03 Draft Idea Paper Complete Draft Proposal Advisor Selection Complete Dissertation First Paper at Proposal Pace Day Final Manuscript and Paper Committee Formation Final Draft Chap 1-3 Draft Chap 1&2 23 Shape Contexts • Shape Contexts are a novel shape descriptor introduced be Belongie[1] • Describes a shape by quantifying each point on the boundary of a shape by its relationship to all the on the boundary points on the shape • Compares shapes by comparing shape contexts [1] S. Belongie, "Image segmentation and shape matching for object recognition," vol. PhD, 2000, pp. 60. 24 Shape Contexts-Constructing x 3 xx 4 5 xx r 6 xx x 2 1 1 xx log r 0 30 60 90 120 150 180 210 Angle in Degrees 240 270 300 330 360 25 Shape Contexts-Constructing 30deg 2 1.732 60deg 90deg 1 Point1 Point 1 2 3 x 0 1 1 y 0 1.732 0 angle 0 60 0 turning angle 270 180 60 adj angle 0 240 300 Point 2 distance 0.0000 2.0000 1.0000 normalized distance 0.0000 1.2680 0.6340 angle 240 0 270 turning angle 270 180 60 adj angle 330 0 210 Point 3 distance 2.0000 0.0000 1.7320 normalized distance 1.2680 0.0000 1.0981 angle 180 90 0 turning angle 270 180 60 adj angle 270 270 0 distance 1.0000 1.7320 0.0000 26 normalized distance 0.6340 1.0981 0.0000 Shape Contexts-Comparing 90 80 70 60 50 40 30 20 10 0 Half Size Triangle Full Size Triangle 90 80 70 60 50 40 30 20 10 0 Count 1 Bin 1 1 Bin 3 2 Bin 5 3 …Bin 60 90 80 70 60 50 40 30 20 10 0 0 Known 2 4 CHI^2 Test where K= # of bins, g is unknown histogram and h is the 2 0 4 known histogram 2 Count Bin 1 Bin 3 Bin 5 Unknown …Bin 60 Count Bin 1 3 4 4 90 80 70 60 50 40 30 20 10 0 0 Bin 3 Bin 5 …Bin 60 Known Count Bin 1 Bin 3 Bin 5 …Bin 60 Known 27 Shape Contexts-Properties: Translation x x x x x x Points X 1 0 Y 0 2 1 1.7321 3 2 3.4642 4 2 1.7321 5 2 0 6 1 0 x x x x x x Points X 1 1 Y 1 2 2 2.7321 3 3 4.4642 4 3 2.7321 5 3 1 6 2 1 28 Shape Contexts-Properties: Scale x x x x x x x x x x x x Points X Y Points X 1 0 0 1 0 Y 0 2 1 1.7321 2 .5 0.866 3 2 3.4642 3 1 1.7321 4 2 1.7321 4 1 0.866 5 2 0 5 1 0 6 1 0 6 .5 0 Point 1 2 3 4 5 6 Point 1 0 6 10 10 10 4 2 6 0 6 6 10 6 3 10 6 0 6 10 10 4 10 6 6 0 6 6 5 10 10 10 6 0 6 6 4 6 10 6 6 0 29 Shape Contexts-Properties: Rotation x x x x x x Points X 1 0 Y 0 2 1 1.7321 3 2 3.4642 4 2 1.7321 5 2 0 6 1 0 x 4 x x 2 x x x 3.4642 X 1 0 Y 0 2 0 1.0 3 0 2.0 4 1.7321 1.0 5 3.4642 0 6 1.7321 0 Point Points 1 2 3 4 5 6 1 6.00 8.00 8.00 6.00 0.00 8.00 2 4.00 6.00 6.00 10.00 8.00 0.00 Point 3 4 0.00 8.00 8.00 0.00 3.33 6.00 10.00 10.00 6.00 8.00 4.00 6.00 5 3.33 6.00 0.00 10.00 8.00 6.00 6 10.00 10.00 10.00 0.00 6.00 10.00 30 Progress Report The Role of Semantic Features in Automobile Identification as of December 10th , 2005 Carl E. Abrams 31 Agenda • Review of Topic and Approach (Elevator Pitch) • Summary Results to Date • Next Steps 32 Review of Topic and Approach • Develop an image segmentation and feature extraction / classification scheme for automobiles that employs the shapes of “semantic” parts and their geometric relationships. – Semantic features are the shapes of : windows, doors, front and rear quarter panels. 33 Review of Topic and Approach • Approach: Develop a test database of vehicles by collecting side images Develop/beg borrow or steal/software to interactively extract the shape and geometric information from the images Work through all the classification test database In parallel continue building the master database – Develop, test and compare segmentation / extraction method for semantic shapes 34 Preliminary Results-Test DB As of Oct 15: All Ford models back to 1990 ~ 50 images As of Nov 12: Acura, Audi, Chrylser, Dodge, Ford, Honda, Mercury, Nissan, Pontiac, Saab, Saturn, Toyota, Volvo, VW models back to 1990 ~125 images 35 Preliminary Results-Image Segmentation Software • Modified CTMRedit: a matlab GUI for viewing, segmenting, and interpolating CT and MRI Images – Written by Mark Hasegawa-Johnson and Jul Cha • Simplified GUI and added capability to store shapes specific to vehicle identification 36 Preliminary Results-Image Segmentation Software Examples 37 Preliminary Results • Create a feature vector that allows the comparison of one shape to another: Vector_Known(ws1,ws2,ws3,ws4,ds1,ds2,body shape) Vector_Unknown(ws1,ws2,ws3,ws4,ds1,ds2,body shape) Feature vector depends on shape descriptor in this case “Shape Contexts” 38 Preliminary Results – Shape Contexts • How to effectively describe a shape? r o 39 Preliminary Results – Shape Contexts • How to effectively describe a shape? r o r (5 bins) O (12 Bins) 40 Preliminary Results – Shape Contexts • How to effectively describe a shape? 90 80 70 60 50 40 30 20 10 0 Count Bin 1 Bin 3 Bin 5 …Bin 60 Develop the Shape Context histograms for every point on the shape 41 Preliminary Results – Distance Between Shape Contexts • How to effectively describe a shape? 90 80 70 60 50 40 30 20 10 0 Count Bin 1 Bin 3 Bin 5 …Bin 60 90 80 70 60 50 40 30 20 10 0 Known 90 80 70 60 50 40 30 20 10 0 CHI^2 Test where K= # of bins, g is unknown histogram and h is the known histogram Count Count Bin 1 Bin 1 Bin 3 Bin 5 …Bin 60 Bin 3 Bin 5 …Bin 60 Known 90 80 70 60 50 40 30 20 10 0 Each shape has 128 points, creates a 128x128 cost matrix Unknown Count Bin 1 Bin 3 Bin 5 …Bin 60 Known 42 Preliminary Results – Distance Between Shape Contexts • What is the best fit (minimum cost) to align all the points? • The Assignment Problem – Hungarian Method for bi-partite matching problem We will be working with the following problem: assign n = 9 candidates to n=9 jobs to minimize the total salary cost paid by the department. The individual salaries of each candidate at each job position depend on their qualification and are given by the cost matrix (in $ per hour): Sa m Jill John Liz Ann Lois Pete Alex Herb Administrator 20 15 10 10 17 23 25 5 15 Secretary to the Chair 10 10 12 15 9 7 8 7 8 Undergraduate Secretary 12 9 9 10 10 5 7 13 9 Graduate Secretary 13 14 10 15 15 5 8 20 10 Financial Clerk 12 13 10 15 14 5 9 20 10 Secretary 15 14 15 16 15 5 10 20 10 Web designer 7 9 12 12 7 6 7 15 12 Receptionist 5 6 8 8 5 4 5 10 7 Typist 5 6 8 8 5 4 5 10 7 If we start with the position of Administrator and assign it to Alex (he gets the minimal salary for this position), then we assign the position of Secretary to Chair to Lois (he gets the minimal salary for this position), and so on, up to the position of Typist, then the assignment is given by the assignment matrix: 43 Preliminary Results • Experimental Setup – Run 5 test cars against known database of 50 cars • Test cars re-segmented from known database • Plot out matches based on Euclidean Distance • Repeat by adding more cars of a different manufacturer to known DB 44 Preliminary Results 300 Unknown: 2003FordMustang2DGT 250 200 Series1 150 100 50 0 2003FordFocus4DSE 2000FordCrownVictoria4D 2004FordCrownVictoriaSedan4D 2003FordCrownVictoria4D 2003FordFocus4DZTS 2001FordCrownVictoria4D 2000FordContour4D 2003FordTaurus4dSES 2002FordCrownVictoria4D 2004FordCrownVictoria4DLX 2002FordFocus4DZTS 2000FordFocus4DLX 2001FordFocus4DLX 2005FordCrownVictoria4D 2001FordFocus4DSE 2000FordTaurus4DSE 2005FordTaurusSE4D 2001FordTaurus4DSES 2005FordFiveHundred4DSE 2001FordEscort4DSedan 2004FordFocus4DZTW 2005FordFocusZX54D 2002FordTaurus4DSES 2002FordEscort4DSedan 2002FordFocus4Dwag 2004FordFocus4DZX5 2002FordFocus4DSE 2004FordTaurus4DSE 2004FordFocus4DZTS 2004FordFocus4DSVT 2003FordFocus4DZX5 2000FordFocus4Dwag 2003FordFoucs4Dwag 2001FordFocus4dwag 2004FordFocus4Dwag 2002FordFocus4DZX5 2001FordMustang2DGT 2000FordEscort4DSedan 2003FordFocusZX3 2005FordFocusZX32D 2002FordFocus2DZX3 2004FordFoucs2DZX3 2003FordFocus2DZX3 2000FordFocus2DZX3 2001FordFocus2DZX3 2005FordMustangCoupe2DV6Deluxe 2003FordMustang2DGT 2000FordMustang2DGT 2004FordMustang2D 2002FordMustang2DGT 45 Preliminary Results Unknown: 2003FordMustang2DGT-Volvos Added to Known DB 300 250 200 Series1 150 100 50 0 2003FordFocus4DSE 2000FordCrownVictoria4D 2004FordCrownVictoriaSedan4D 2000VolvoS704D 2003VolvoS804D 2003FordCrownVictoria4D 2003FordFocus4DZTS 2001FordCrownVictoria4D 2000FordContour4D 2003FordTaurus4dSES 2002FordCrownVictoria4D 2004FordCrownVictoria4DLX 2002FordFocus4DZTS 2000FordFocus4DLX 2005VolvoS804D 2001FordFocus4DLX 2005FordCrownVictoria4D 2003VolvoS404D 2002VolvoS804D 2001VolvoS404D 2001FordFocus4DSE 2000VolvoS804D 2001VolvoS804D 2004VolvoS804D 2000FordTaurus4DSE 2005FordTaurusSE4D 2001FordTaurus4DSES 2000VolvoS404D 2005FordFiveHundred4DSE 2001FordEscort4DSedan 2004FordFocus4DZTW 2005FordFocusZX54D 2002FordTaurus4DSES 2002FordEscort4DSedan 2002FordFocus4Dwag 2004VolvoS604D 2004FordFocus4DZX5 2002FordFocus4DSE 2004FordTaurus4DSE 2004FordFocus4DZTS 2005VolvoS404D 2004FordFocus4DSVT 2003FordFocus4DZX5 2001VolvoS604D 2002VolvoS604D 2000FordFocus4Dwag 2005VolvoS604D 2003VolvoS604D 2003FordFoucs4Dwag 2001FordFocus4dwag 2004FordFocus4Dwag 2002FordFocus4DZX5 2001FordMustang2DGT 2000FordEscort4DSedan 2004VolvoS404D 2003FordFocusZX3 2005FordFocusZX32D 2002FordFocus2DZX3 2004FordFoucs2DZX3 2003FordFocus2DZX3 2000FordFocus2DZX3 2001FordFocus2DZX3 2005FordMustangCoupe2DV6De 2003FordMustang2DGT 2000FordMustang2DGT 2004FordMustang2D 2002FordMustang2DGT 46 Preliminary Results Unknown: 2004FordFocus4DZX5.dh1 300 250 200 Series1 150 100 50 0 2003FordFocusZX3 2005FordMustangCoupe2DV6Deluxe 2004FordMustang2D 2005FordFocusZX32D 2002FordMustang2DGT 2000FordMustang2DGT 2001FordFocus2DZX3 2003FordMustang2DGT 2002FordFocus2DZX3 2004FordFoucs2DZX3 2001FordMustang2DGT 2003FordFocus2DZX3 2000FordFocus2DZX3 2000FordEscort4DSedan 2002FordFocus4DZX5 2004FordCrownVictoriaSedan4D 2005FordCrownVictoria4D 2003FordCrownVictoria4D 2001FordCrownVictoria4D 2000FordCrownVictoria4D 2002FordCrownVictoria4D 2004FordCrownVictoria4DLX 2005FordTaurusSE4D 2000FordContour4D 2002FordEscort4DSedan 2000FordTaurus4DSE 2003FordTaurus4dSES 2005FordFiveHundred4DSE 2001FordEscort4DSedan 2003FordFocus4DZTS 2005FordFocusZX54D 2003FordFocus4DSE 2004FordFocus4DZTS 2001FordTaurus4DSES 2002FordFocus4DZTS 2002FordTaurus4DSES 2002FordFocus4DSE 2001FordFocus4DLX 2004FordTaurus4DSE 2001FordFocus4DSE 2000FordFocus4Dwag 2004FordFocus4DZTW 2003FordFoucs4Dwag 2002FordFocus4Dwag 2004FordFocus4Dwag 2001FordFocus4dwag 2000FordFocus4DLX 2004FordFocus4DSVT 2003FordFocus4DZX5 2004FordFocus4DZX5 47 Preliminary Results Unknown: 2004FordFocus4DZX5.dh1-Volvos Added to Known DB 300 250 200 Series1 150 100 50 0 2003FordFocusZX3 2005FordMustangCoupe2DV6D 2004FordMustang2D 2005FordFocusZX32D 2002FordMustang2DGT 2000FordMustang2DGT 2001FordFocus2DZX3 2003FordMustang2DGT 2002FordFocus2DZX3 2004FordFoucs2DZX3 2001FordMustang2DGT 2003FordFocus2DZX3 2000FordFocus2DZX3 2004VolvoS404D 2000FordEscort4DSedan 2002FordFocus4DZX5 2004FordCrownVictoriaSedan4D 2005FordCrownVictoria4D 2003FordCrownVictoria4D 2001FordCrownVictoria4D 2000FordCrownVictoria4D 2002FordCrownVictoria4D 2003VolvoS604D 2004FordCrownVictoria4DLX 2003VolvoS804D 2005FordTaurusSE4D 2000VolvoS704D 2000FordContour4D 2002FordEscort4DSedan 2002VolvoS604D 2000VolvoS804D 2001VolvoS804D 2004VolvoS604D 2005VolvoS804D 2001VolvoS404D 2001VolvoS604D 2000FordTaurus4DSE 2003FordTaurus4dSES 2005FordFiveHundred4DSE 2001FordEscort4DSedan 2003FordFocus4DZTS 2005FordFocusZX54D 2002VolvoS804D 2003FordFocus4DSE 2005VolvoS604D 2004VolvoS804D 2000VolvoS404D 2004FordFocus4DZTS 2001FordTaurus4DSES 2002FordFocus4DZTS 2003VolvoS404D 2002FordTaurus4DSES 2002FordFocus4DSE 2001FordFocus4DLX 2004FordTaurus4DSE 2001FordFocus4DSE 2000FordFocus4Dwag 2004FordFocus4DZTW 2005VolvoS404D 2003FordFoucs4Dwag 2002FordFocus4Dwag 2004FordFocus4Dwag 2001FordFocus4dwag 2000FordFocus4DLX 2004FordFocus4DSVT 2003FordFocus4DZX5 2004FordFocus4DZX5 48 Preliminary Results Unknown: 2004FordTaurus4DSES.dh1 300 250 200 Series1 150 100 50 0 2003FordFocusZX3 2005FordFocusZX32D 2005FordMustangCoupe2DV6Deluxe 2002FordMustang2DGT 2004FordMustang2D 2003FordMustang2DGT 2000FordMustang2DGT 2004FordFoucs2DZX3 2001FordFocus2DZX3 2002FordFocus2DZX3 2003FordFocus2DZX3 2000FordFocus2DZX3 2001FordMustang2DGT 2002FordFocus4DZX5 2000FordEscort4DSedan 2004FordCrownVictoriaSedan4D 2004FordFocus4DZTW 2003FordCrownVictoria4D 2003FordFoucs4Dwag 2001FordCrownVictoria4D 2000FordCrownVictoria4D 2005FordCrownVictoria4D 2002FordCrownVictoria4D 2004FordCrownVictoria4DLX 2002FordFocus4Dwag 2001FordFocus4dwag 2001FordEscort4DSedan 2005FordFocusZX54D 2000FordFocus4Dwag 2002FordEscort4DSedan 2004FordFocus4DZTS 2000FordContour4D 2004FordFocus4DSVT 2002FordFocus4DSE 2005FordFiveHundred4DSE 2003FordFocus4DZX5 2004FordFocus4DZX5 2003FordFocus4DSE 2004FordFocus4Dwag 2003FordFocus4DZTS 2002FordFocus4DZTS 2001FordFocus4DLX 2000FordFocus4DLX 2001FordFocus4DSE 2003FordTaurus4dSES 2000FordTaurus4DSE 2001FordTaurus4DSES 2005FordTaurusSE4D 2004FordTaurus4DSE 2002FordTaurus4DSES 49 Preliminary Results Unknown: 2004FordTaurus4DSES.dh1-Volvos Added to Known DB 300 250 200 Series1 150 100 50 0 2003FordFocusZX3 2005FordFocusZX32D 2005FordMustangCoupe2DV6Delux 2002FordMustang2DGT 2004FordMustang2D 2003FordMustang2DGT 2000FordMustang2DGT 2004FordFoucs2DZX3 2001FordFocus2DZX3 2002FordFocus2DZX3 2003FordFocus2DZX3 2000FordFocus2DZX3 2001FordMustang2DGT 2004VolvoS404D 2002FordFocus4DZX5 2000FordEscort4DSedan 2004FordCrownVictoriaSedan4D 2004FordFocus4DZTW 2003FordCrownVictoria4D 2003FordFoucs4Dwag 2001FordCrownVictoria4D 2000FordCrownVictoria4D 2005FordCrownVictoria4D 2002FordCrownVictoria4D 2004FordCrownVictoria4DLX 2002FordFocus4Dwag 2001FordFocus4dwag 2001FordEscort4DSedan 2005FordFocusZX54D 2000FordFocus4Dwag 2002FordEscort4DSedan 2004FordFocus4DZTS 2000VolvoS704D 2000FordContour4D 2003VolvoS604D 2004FordFocus4DSVT 2002VolvoS604D 2002FordFocus4DSE 2001VolvoS604D 2004VolvoS604D 2005FordFiveHundred4DSE 2003FordFocus4DZX5 2004FordFocus4DZX5 2003FordFocus4DSE 2005VolvoS604D 2004FordFocus4Dwag 2003FordFocus4DZTS 2002FordFocus4DZTS 2001FordFocus4DLX 2000FordFocus4DLX 2003VolvoS804D 2000VolvoS804D 2001FordFocus4DSE 2001VolvoS804D 2002VolvoS804D 2001VolvoS404D 2005VolvoS804D 2005VolvoS404D 2003FordTaurus4dSES 2000VolvoS404D 2003VolvoS404D 2000FordTaurus4DSE 2004VolvoS804D 2001FordTaurus4DSES 2005FordTaurusSE4D 2004FordTaurus4DSE 2002FordTaurus4DSES 50 Preliminary Results Unknown: 2005FordMustangCoupe2DV6Deluxe.dh1 300 250 200 Series1 150 100 50 0 2000FordCrownVictoria4D 2003FordCrownVictoria4D 2002FordCrownVictoria4D 2004FordCrownVictoriaSedan4D 2003FordFocus4DSE 2001FordCrownVictoria4D 2004FordCrownVictoria4DLX 2002FordFocus4DZTS 2000FordContour4D 2005FordCrownVictoria4D 2001FordEscort4DSedan 2001FordFocus4DLX 2000FordFocus4DLX 2003FordFocus4DZTS 2000FordTaurus4DSE 2002FordEscort4DSedan 2001FordFocus4DSE 2001FordTaurus4DSES 2004FordFocus4DZTW 2004FordFocus4DZTS 2003FordTaurus4dSES 2004FordFocus4DSVT 2002FordFocus4DSE 2004FordFocus4DZX5 2002FordFocus4Dwag 2002FordTaurus4DSES 2003FordFoucs4Dwag 2003FordFocus4DZX5 2004FordTaurus4DSE 2001FordFocus4dwag 2000FordFocus4Dwag 2005FordTaurusSE4D 2004FordFocus4Dwag 2005FordFiveHundred4DSE 2005FordFocusZX54D 2001FordMustang2DGT 2002FordFocus4DZX5 2000FordEscort4DSedan 2003FordMustang2DGT 2000FordMustang2DGT 2002FordFocus2DZX3 2003FordFocus2DZX3 2000FordFocus2DZX3 2002FordMustang2DGT 2004FordFoucs2DZX3 2001FordFocus2DZX3 2005FordFocusZX32D 2003FordFocusZX3 2004FordMustang2D 2005FordMustangCoupe2DV6Deluxe 51 Preliminary Results Unknown: 2005FordMustangCoupe2DV6Deluxe.dh1-Volvos Added to Known DB 300 250 200 Series1 150 100 50 0 2000FordCrownVictoria4D 2003FordCrownVictoria4D 2002FordCrownVictoria4D 2004FordCrownVictoriaSedan4D 2003FordFocus4DSE 2000VolvoS704D 2001FordCrownVictoria4D 2004FordCrownVictoria4DLX 2001VolvoS404D 2002FordFocus4DZTS 2000FordContour4D 2005FordCrownVictoria4D 2003VolvoS804D 2003VolvoS404D 2001FordEscort4DSedan 2001VolvoS804D 2001FordFocus4DLX 2000FordFocus4DLX 2000VolvoS404D 2003FordFocus4DZTS 2000FordTaurus4DSE 2002FordEscort4DSedan 2000VolvoS804D 2001FordFocus4DSE 2002VolvoS804D 2005VolvoS804D 2001FordTaurus4DSES 2004FordFocus4DZTW 2004FordFocus4DZTS 2003FordTaurus4dSES 2004FordFocus4DSVT 2002FordFocus4DSE 2004FordFocus4DZX5 2002FordFocus4Dwag 2001VolvoS604D 2004VolvoS804D 2002FordTaurus4DSES 2003FordFoucs4Dwag 2003FordFocus4DZX5 2004FordTaurus4DSE 2004VolvoS604D 2003VolvoS604D 2005VolvoS604D 2001FordFocus4dwag 2000FordFocus4Dwag 2002VolvoS604D 2005FordTaurusSE4D 2005VolvoS404D 2004FordFocus4Dwag 2005FordFiveHundred4DSE 2005FordFocusZX54D 2001FordMustang2DGT 2002FordFocus4DZX5 2000FordEscort4DSedan 2004VolvoS404D 2003FordMustang2DGT 2000FordMustang2DGT 2002FordFocus2DZX3 2003FordFocus2DZX3 2000FordFocus2DZX3 2002FordMustang2DGT 2004FordFoucs2DZX3 2001FordFocus2DZX3 2005FordFocusZX32D 2003FordFocusZX3 2004FordMustang2D 2005FordMustangCoupe2DV6Deluxe 52 Preliminary Results Unknown: 2005FordTaurusSE4D.ws2 300 250 200 Series1 150 100 50 0 2003FordFocusZX3 2005FordFocusZX32D 2005FordMustangCoupe2DV6Delu 2004FordFoucs2DZX3 2002FordFocus2DZX3 2001FordFocus2DZX3 2003FordFocus2DZX3 2002FordMustang2DGT 2000FordFocus2DZX3 2003FordMustang2DGT 2000FordMustang2DGT 2004FordMustang2D 2001FordMustang2DGT 2002FordFocus4DZX5 2000FordEscort4DSedan 2004FordFocus4DZTW 2001FordCrownVictoria4D 2003FordFoucs4Dwag 2003FordCrownVictoria4D 2002FordCrownVictoria4D 2004FordCrownVictoria4DLX 2002FordFocus4Dwag 2000FordCrownVictoria4D 2000FordFocus4Dwag 2001FordEscort4DSedan 2001FordFocus4dwag 2004FordCrownVictoriaSedan4D 2004FordFocus4DZTS 2002FordEscort4DSedan 2000FordContour4D 2004FordFocus4DSVT 2003FordFocus4DZX5 2005FordCrownVictoria4D 2002FordFocus4DSE 2004FordFocus4DZX5 2004FordFocus4Dwag 2001FordFocus4DLX 2002FordFocus4DZTS 2003FordFocus4DZTS 2005FordFiveHundred4DSE 2003FordFocus4DSE 2005FordFocusZX54D 2001FordFocus4DSE 2000FordFocus4DLX 2001FordTaurus4DSES 2003FordTaurus4dSES 2000FordTaurus4DSE 2004FordTaurus4DSE 2002FordTaurus4DSES 2005FordTaurusSE4D 53 Preliminary Results Unknown: 2005FordTaurusSE4D.ws2-Vovlos Added to Known DB 300 250 200 Series1 150 100 50 0 2003FordFocusZX3 2005FordFocusZX32D 2005FordMustangCoupe2DV6Deluxe 2004FordFoucs2DZX3 2002FordFocus2DZX3 2001FordFocus2DZX3 2003FordFocus2DZX3 2002FordMustang2DGT 2000FordFocus2DZX3 2003FordMustang2DGT 2000FordMustang2DGT 2004FordMustang2D 2001FordMustang2DGT 2004VolvoS404D 2002FordFocus4DZX5 2000FordEscort4DSedan 2004FordFocus4DZTW 2001FordCrownVictoria4D 2003FordFoucs4Dwag 2003FordCrownVictoria4D 2002FordCrownVictoria4D 2004FordCrownVictoria4DLX 2002FordFocus4Dwag 2000FordCrownVictoria4D 2000FordFocus4Dwag 2001FordEscort4DSedan 2001FordFocus4dwag 2004FordCrownVictoriaSedan4D 2004FordFocus4DZTS 2000VolvoS704D 2001VolvoS604D 2002FordEscort4DSedan 2000FordContour4D 2004FordFocus4DSVT 2003FordFocus4DZX5 2005FordCrownVictoria4D 2002FordFocus4DSE 2004FordFocus4DZX5 2005VolvoS604D 2004FordFocus4Dwag 2004VolvoS604D 2002VolvoS604D 2001FordFocus4DLX 2002FordFocus4DZTS 2003FordFocus4DZTS 2003VolvoS604D 2005FordFiveHundred4DSE 2003FordFocus4DSE 2005FordFocusZX54D 2001FordFocus4DSE 2000FordFocus4DLX 2000VolvoS404D 2000VolvoS804D 2001VolvoS404D 2005VolvoS404D 2001VolvoS804D 2003VolvoS404D 2003VolvoS804D 2001FordTaurus4DSES 2002VolvoS804D 2005VolvoS804D 2003FordTaurus4dSES 2000FordTaurus4DSE 2004FordTaurus4DSE 2004VolvoS804D 2002FordTaurus4DSES 2005FordTaurusSE4D 54 Known vs Known (as of Nov 12th) 2003FordMustang2DGT 254.185 250.633 109.297 249.126 108.027 254.113 251.303 259.786 243.888 245.835 242.243 242.393 241.877 2003FordTaurus4dSES 135.555 126.965 2002FordCrownVictoria4D 103.196 145.769 266.257 154.208 265.434 145.884 2002FordEscort4DSedan 116.005 137.942 259.742 146.017 257.911 137.572 118.731 2002FordFocus2DZX3 258.637 245.807 2002FordFocus4DSE 120.559 125.648 253.573 136.145 256.645 133.528 119.804 138.053 101.793 77.909 114.632 103.044 103.852 246.468 252.282 122.402 122.582 243.567 122.053 105.311 107.885 261.017 107.077 245.769 123.098 115.344 103.025 244.644 2002FordFocus4DZTS 123.436 132.637 257.362 95.894 122.939 2002FordFocus4DZX5 198.039 185.814 193.622 190.509 2002FordFocus4Dwag 145.92 126.825 249.754 133.968 254.776 142.713 136.715 159.707 108.293 115.939 2002FordMustang2DGT 252.749 253.154 115.616 249.333 2002FordTaurus4DSES 138.034 120.629 253.084 125.365 253.797 2001FordCrownVictoria4D 105.877 145.624 265.211 155.231 264.963 148.372 2001FordEscort4DSedan 113.581 142.068 257.541 146.427 260.258 143.145 107.663 130.771 2001FordFocus2DZX3 259.965 246.122 257.07 131.378 256.423 103.991 140.199 141.583 126.55 129.347 134.36 120.282 102.844 96.471 83.914 98.935 142.51 125.311 138.6 260.929 129.964 120.872 144.701 100.442 78.302 247.309 200.62 197.209 93.433 84.64 120.67 127.234 255.575 132.305 259.787 128.922 119.559 116.7 141.16 101.628 244.88 253.72 96.584 256.169 95.446 253.432 250.701 243.432 96.503 134.327 250.036 117.194 110.082 129.097 121.63 121.63 240.197 59.413 110.627 245.89 258.902 58.859 252.409 248.886 237.613 107.298 239.085 101.02 113.555 248.983 257.252 118.082 115.731 247.119 114.23 82.862 109.322 264.954 114.336 75.821 247.53 248.699 90.601 88.533 240.213 250.831 123.233 143.025 240.064 120.592 127.662 248.03 242.277 105.535 78.079 245.851 254.476 105.698 242.134 213.063 244.379 208.178 235.636 240.249 242.75 129.59 252.44 248.911 239.31 237.961 52.756 252.644 248.256 238.632 100.894 239.646 196.54 191.566 76.397 254.313 90.21 250.296 259.528 253.495 97.877 109.601 248.279 256.185 117.737 120.782 246.064 114.37 88.679 112.068 263.728 98.903 105.584 247.511 254.727 114.624 127.366 244.459 109.16 86.645 104.156 262.256 113.026 249.738 114.666 125.043 112.189 89.081 237.54 241.778 238.597 238.716 205.247 198.905 232.034 115.44 250.068 252.358 96.834 245.101 123.336 82.491 238.716 86.289 242.791 120.928 109.932 110.964 125.418 185.707 120.062 118.034 118.608 125.032 134.287 121.705 124.786 130.423 120.302 129.092 118.129 134.876 126.309 88.533 242.277 112.643 139.877 127.119 237.961 109.601 105.584 115.26 250.378 113.234 115.325 110.263 247.308 54.542 248.279 247.511 237.559 205.247 250.068 252.011 260.733 180.198 57.029 246.362 237.721 78.079 250.236 260.689 256.082 107.539 256.185 254.727 246.852 198.905 252.358 257.427 263.675 191.438 107.726 256.634 247.195 74.022 136.062 127.837 247.457 117.737 114.624 121.02 232.034 98.61 109.502 260.054 120.782 127.366 137.549 95.357 242.886 134.861 134.481 125.304 239.646 196.54 136.187 76.397 246.215 252.921 244.445 0 97.054 248.72 251.254 261.499 180.221 98.59 241.823 120.105 135.265 46.894 245.935 235.991 144.97 172.544 235.817 105.617 246.81 90.21 250.378 249.738 238.753 192.131 248.709 250.832 255.242 183.432 99.03 253.495 110.263 112.189 125.847 236.233 114.165 108.472 114.508 63.39 247.308 246.81 237.591 208.367 250.063 251.167 260.2 179.686 55.927 246.183 237.501 104.865 252.938 244.345 243.724 253.856 183.192 248.881 193.28 185.406 195.87 185.257 180.887 180.802 178.444 167.249 260.893 188.247 190.578 196.504 0 247.954 124.154 139.139 123.242 193.28 247.954 240.59 119.026 0 248.889 254.634 248.929 104.419 0 139.078 195.87 139.139 254.634 139.078 95.798 107.142 185.257 123.242 248.929 252.72 115.78 252.68 241.289 193.691 128.81 247.562 119.405 115.554 126.239 232.557 128.81 102.576 69.019 180.802 119.026 252.72 119.405 119.407 105.013 248.686 0 79.566 178.444 252.68 115.554 129.368 110.792 247.067 77.55 77.55 105.867 235.369 110.645 105.268 0 108.385 87.677 241.289 126.239 133.164 119.436 237.736 105.867 108.385 236.68 234.904 260.893 240.793 193.691 232.557 249.72 242.06 235.053 204.352 235.369 234.22 107.418 111.678 57.034 247.419 238.534 0 98.423 2000FordTaurus4DSE 140.154 125.609 259.223 137.202 258.295 253.55 252.555 91.873 137.231 251.729 118.983 124.483 123.458 263.481 140.966 253.757 100.392 138.951 130.477 252.938 130.37 121.476 191.263 135.152 253.189 91.051 2005VolvoS404D 139.568 108.818 108.94 109.932 242.512 245.697 92.815 143.699 243.968 113.065 119.475 110.802 254.774 131.535 243.202 103.697 136.726 134.506 244.345 124.29 119.451 180.523 118.119 246.443 91.142 141.857 131.614 243.069 116.239 2005VolvoS604D 114.853 131.146 251.332 131.953 251.525 131.138 112.934 132.239 107.195 105.576 123.934 110.488 110.964 243.536 248.713 113.907 123.538 2005VolvoS804D 131.257 122.848 2004VolvoS404D 203.158 2004VolvoS604D 109.743 136.232 253.497 134.602 253.439 129.768 115.864 2004VolvoS804D 132.665 121.666 256.946 129.588 256.318 101.555 136.978 143.441 121.793 122.669 2003VolvoS404D 134.759 122.998 259.452 135.124 261.371 108.446 129.094 142.877 122.418 124.086 126.276 118.477 118.608 2003VolvoS604D 108.281 135.307 254.441 135.967 251.386 129.381 114.761 131.203 118.334 106.241 142.173 128.289 125.032 248.957 248.194 125.749 119.556 248.211 144.338 120.128 136.251 2003VolvoS804D 137.478 135.795 262.649 144.006 260.925 117.179 143.779 143.962 138.412 133.004 150.106 138.319 134.287 256.576 257.205 109.328 144.085 257.215 107.338 123.244 2002VolvoS604D 115.765 137.754 254.006 131.602 250.662 127.275 116.341 129.589 118.825 121.398 136.539 122.878 121.705 248.365 246.729 119.084 126.832 248.541 143.352 128.932 137.072 258.16 132.326 2002VolvoS804D 119.794 126.436 259.622 129.777 258.429 112.737 264.21 133.727 250.474 107.067 120.104 120.512 251.881 98.512 140.734 148.023 125.431 133.33 140.309 123.336 120.928 249.19 115.675 249.251 105.521 138.868 148.277 117.111 131.297 125.731 260.64 136.571 258.306 107.819 135.753 139.967 127.206 122.855 194.59 190.601 197.758 191.282 133.11 122.952 103.68 255.056 145.44 128.956 125.418 255.001 254.834 103.039 137.649 253.598 107.402 110.377 134.06 119.575 120.062 109.92 118.231 164.587 69.287 246.602 253.937 247.388 129.61 265.075 141.959 254.605 102.798 132.605 123.477 253.856 116.552 112.507 194.814 139.637 255.142 105.337 252.75 255.853 95.405 139.664 250.064 106.992 111.167 124.838 263.196 134.866 250.051 98.952 134.356 127.853 250.34 115.648 112.052 189.964 131.144 251.91 89.416 130.383 252.315 104.795 120.119 118.228 265.924 130.783 250.719 102.934 127.896 126.473 252.677 118.918 110.887 185.965 122.959 253.067 257.82 130.351 246.346 126.967 113.985 102.236 143.07 267.242 115.76 124.786 253.584 254.148 105.744 126.922 252.673 114.359 108.278 126.805 93.925 137.891 127.001 250.838 110.46 0 0 192.51 0 79.567 120.648 191.94 231.12 103.514 118.123 125.766 185.245 250.433 112.64 113.63 123.025 132.8 98.488 125.251 249.419 94.601 113.35 128.594 248.8 133.557 131.062 244.198 133.519 118.659 140.47 252.721 102.522 129.28 247.674 130.485 121.207 2001VolvoS804D 132.043 130.283 264.453 143.188 261.362 2000VolvoS404D 134.073 131.215 258.266 133.489 259.834 105.623 131.724 142.322 122.008 123.792 134.135 121.766 118.129 252.189 252.962 2000VolvoS704D 116.462 143.599 264.866 149.439 265.884 138.436 110.593 129.603 120.475 110.508 137.509 126.692 134.876 258.149 261.667 129.656 108.974 256.925 130.427 2000VolvoS804D 137.533 125.878 260.986 139.069 258.365 114.913 143.755 144.485 134.051 129.795 147.934 135.003 126.309 255.34 122.6 257.576 121.282 114.45 136.117 269.766 131.687 254.67 107.246 143.059 254.573 107.367 117.258 254.98 123.734 103.019 114.793 256.581 112.539 137.97 265.497 145.036 253.958 118.1 119.567 234.143 120.049 108.958 117.125 189.006 249.228 132.634 116.28 196.753 143.672 253.035 107.218 135.882 132.231 258.131 116.593 115.777 137.268 230.627 107.568 131.34 252.207 117.098 119.875 125.153 264.962 134.456 249.801 104.451 131.101 124.025 253.513 119.237 111.002 190.915 130.305 249.689 93.127 133.748 122.129 253.553 113.492 97.594 194.904 137.951 255.861 133.284 109.717 110.332 258.856 106.26 138.149 125.836 255.608 125.377 120.622 196.225 142.665 252.324 106.224 111.56 124.632 228.986 94.796 108.182 127.514 125.54 135.952 198.457 256.38 96.067 139.14 135.504 255.604 119.683 117.015 138.174 229.169 107.955 131.047 98.664 194.584 255.413 115.26 126.7 192.51 120.648 113.63 112.64 122.278 106.945 105.831 120.049 107.568 132.8 119.451 113.496 115.694 108.958 88.814 248.8 252.162 254.656 249.228 99.626 128.844 78.912 102.241 0 119.66 124.81 91.503 91.522 124.153 95.058 124.804 0 132.1 98.664 140.75 140.47 256.38 251.267 255.413 255.031 119.88 113.676 128.28 116.374 129.28 143.925 130.743 134.246 142.607 89.786 98.732 105.686 119.611 109.494 192.066 123.369 100.848 80.053 197.382 121.888 85.794 106.198 84.643 83.597 109.494 119.702 132.1 108.847 131.234 102.772 86.707 126.424 99.884 0 130.935 94.092 119.181 128.205 0 128.868 100.658 105.278 94.092 128.868 94.086 127.76 0 117.668 130.428 127.76 92.195 125.515 101.307 85.794 82.913 119.155 106.198 126.88 122.361 125.33 92.195 111.343 128.146 83.639 86.832 125.515 124.028 129.361 127.368 0 126.626 86.832 113.958 126.626 83.639 127.368 94.086 122.297 0 102.683 113.958 101.307 110.524 126.88 111.343 124.028 110.524 125.33 80.053 123.44 123.369 121.004 121.888 89.786 100.848 123.931 99.884 122.297 82.913 99.73 118.752 190.35 195.694 191.709 196.672 192.066 201.557 197.382 86.707 128.205 105.278 130.428 102.683 84.643 120.162 126.424 77.431 124.645 119.611 123.043 125.786 98.648 120.162 90.385 102.772 119.181 100.658 117.668 98.648 88.814 126.7 89.077 114.207 128.077 92.508 118.706 98.732 133.625 108.042 115.26 121.207 113.541 105.686 134.521 110.738 90.385 92.508 108.847 130.935 89.077 118.706 131.234 123.44 78.957 113.487 192.89 192.631 202.383 194.306 95.058 99.73 195.694 128.077 113.35 118.659 103.222 94.869 128.594 128.28 134.246 255.259 133.625 134.521 123.043 119.702 201.557 121.004 123.931 119.155 122.361 128.146 129.361 140.75 197.027 255.031 116.374 142.607 254.004 108.042 110.738 125.786 96.067 131.047 189.76 192.875 189.006 198.457 190.326 194.584 197.027 94.601 128.672 113.436 133.519 116.028 102.522 130.485 103.889 91.522 124.804 190.35 114.207 83.597 196.672 239.67 229.169 97.814 125.692 107.955 125.54 118.774 137.26 125.286 118.442 127.171 117.125 135.952 129.829 191.84 197.084 188.898 98.488 140.064 121.394 133.557 112.535 120.468 132.634 0 117.857 128.431 91.503 202.383 124.153 124.81 194.306 96.214 191.94 189.034 192.89 128.431 119.66 192.631 77.431 118.752 191.709 119.88 143.925 254.783 103.889 113.541 124.645 97.814 118.774 129.829 190.326 251.267 113.676 130.743 248.985 239.67 125.692 78.957 189.76 252.162 112.535 132.155 250.692 116.028 103.222 113.487 134.19 137.555 263.666 143.532 262.547 109.823 131.254 141.008 128.343 123.283 142.092 128.863 130.423 257.016 255.872 102.362 127.681 256.451 112.089 116.975 130.514 267.976 142.515 254.233 105.984 128.801 120.939 256.571 120.149 111.959 194.555 138.871 253.365 106.766 133.367 122.203 257.352 116.534 109.856 136.112 230.468 105.831 115.694 127.171 192.875 254.656 120.468 90.023 94.869 94.796 119.683 111.56 108.182 117.015 231.12 233.919 227.408 231.965 232.071 230.468 234.143 230.627 228.986 79.567 114.388 123.776 78.912 189.034 117.857 99.626 123.776 102.241 191.84 248.496 140.064 132.904 246.942 128.672 128.844 137.26 197.084 257.117 121.394 147.787 253.893 113.436 99.203 127.675 232.071 106.945 113.496 118.442 96.214 114.388 110.315 132.187 256.627 137.607 255.121 133.779 111.632 125.729 115.016 108.771 134.012 119.397 120.302 249.648 251.846 120.878 121.997 248.974 137.062 122.697 132.838 262.383 128.596 246.139 122.138 107.319 105.979 249.343 258.01 255.888 105.103 135.858 256.814 111.552 117.643 137.309 268.768 143.823 255.241 102.782 133.012 0 122.051 189.171 254.4 107.695 106.179 122.051 246.85 132.676 119.896 118.1 115.777 246.85 250.793 249.419 246.942 253.893 244.198 250.692 252.721 247.674 254.783 248.985 255.259 254.004 0 116.226 106.179 184.308 119.896 2001VolvoS604D 112.22 137.999 141.344 129.514 126.692 147.548 132.567 129.092 254.4 184.341 104.08 129.495 107.695 194.873 132.676 108.033 2001VolvoS404D 103.65 115.992 189.539 131.501 245.822 122.006 104.641 109.221 249.419 107.594 231 231.851 170.452 231.617 230.781 190.63 170.452 190.353 199.419 207.095 238.356 179.958 194.734 192.479 184.341 194.873 184.308 189.171 103.16 128.858 230.781 100.199 126.141 134.189 189.599 249.246 103.647 131.434 250.793 108.033 257.57 124.222 122.894 143.074 227.408 100.92 99.203 109.856 0 252.169 242.633 243.633 104.08 244.21 125.527 118.607 119.162 248.827 116.547 127.942 190.977 133.592 242.556 121.513 116.232 118.418 248.692 121.597 127.112 123.451 231.965 122.278 119.451 125.286 188.898 111.58 103.874 190.303 129.673 253.174 106.057 125.959 116.997 253.255 100.92 116.534 107.594 116.593 113.492 103.16 107.738 120.131 122.894 127.112 0 242.079 136.286 122.307 120.311 138.898 192.479 128.956 131.434 125.251 132.904 147.787 131.062 132.155 97.364 130.148 141.315 189.061 251.927 117.293 136.286 252.169 92.803 132.095 121.663 253.632 111.564 107.738 122.811 150.47 254.599 108.729 138.632 132.779 257.229 130.936 124.667 197.806 145.495 255.297 107.368 141.479 135.314 110.46 111.564 115.595 124.222 121.597 231 115.679 110.124 121.206 181.921 243.412 125.257 120.311 243.633 129.495 116.226 249.98 111.041 118.655 192.712 139.879 245.333 127.973 114.937 114.739 249.401 115.595 120.131 126.397 233.919 121.079 139.14 190.63 119.777 128.858 122.811 126.397 143.074 123.451 127.675 136.112 119.567 137.268 124.632 127.514 138.174 97.364 103.964 115.679 104.523 190.353 119.938 100.199 103.514 121.079 114.85 121.806 231.323 103.964 131.859 137.779 177.418 243.361 100.362 122.307 242.633 187.99 185.822 93.127 133.284 106.224 187.99 109.903 0 112.063 247.805 117.293 100.362 125.257 113.261 194.734 133.788 103.647 132.12 129.828 255.109 110.878 105.279 136.724 231.851 104.523 124.426 133.038 194.387 253.148 113.261 138.898 196.55 206.244 196.366 183.192 190.097 189.368 239.203 194.782 184.153 190.167 208.651 199.026 182.603 97.594 120.622 257.57 248.692 253.255 257.352 249.419 258.131 253.553 258.856 255.604 0 245.073 235.402 100.842 251.927 243.361 243.412 253.148 179.958 247.667 249.246 250.433 248.496 257.117 78.287 242.778 132.302 136.207 141.951 170.992 235.402 112.063 142.61 139.576 253.805 123.575 122.401 135.172 234.247 92.803 127.973 107.368 121.513 106.057 106.766 122.006 107.218 0 178.476 178.445 170.992 183.993 189.061 177.418 181.921 194.387 238.356 189.688 189.599 185.245 99.486 251.893 251.422 240.587 189.266 249.139 252.103 254.495 183.993 100.842 247.805 242.079 99.166 114.637 182.978 121.904 243.733 114.043 112.264 104.258 243.966 108.121 112.993 111.201 93.925 0 193.879 259.611 134.943 141.951 254.495 141.315 137.779 121.206 133.038 207.095 119.637 134.189 125.766 123.025 57.034 246.021 243.971 235.974 206.637 249.613 249.531 259.611 178.476 85.26 244.409 128.888 137.892 129.397 238.534 112.526 111.193 66.089 252.819 253.125 250.173 104.865 248.623 252.249 188.225 242.253 191.99 187.107 198.481 195.943 183.506 99.813 116.28 111.002 114.85 112.993 105.279 185.822 116.077 78.287 240.587 135.172 121.806 111.201 136.724 247.47 250.063 121.717 123.098 247.328 136.531 119.854 133.987 259.392 130.056 245.696 121.669 115.352 110.064 248.881 104.018 117.074 189.491 132.351 246.278 121.677 112.539 108.587 248.388 109.903 116.077 119.777 231.617 119.938 109.456 119.637 189.688 247.667 133.788 128.956 136.79 122.493 118.034 250.389 252.463 124.57 130.835 117.285 114.291 133.772 78.318 241.306 139.581 139.531 134.721 237.501 124.72 135.502 241.497 135.783 71.019 171.218 190.724 184.418 193.412 182.115 181.245 184.765 183.439 173.241 258.675 188.756 186.979 193.879 55.927 242.748 246.758 178.257 239.053 106.166 246.107 256.365 247.805 242.81 133.922 119.262 125.755 258.431 120.786 242.033 120.486 114.526 110.533 243.724 193.55 205.675 215.788 186.813 193.982 198.754 184.521 185.707 182.543 184.347 188.226 206.591 180.389 202.382 130.05 115.397 107.655 94.613 254.137 252.82 244.925 122.205 242.719 116.02 103.65 121.282 119.237 112.539 125.377 251.91 253.067 245.333 255.297 242.556 253.174 253.365 245.822 253.035 249.689 255.861 252.324 99.813 186.979 249.531 130.974 136.207 252.103 130.148 131.859 110.124 124.426 199.419 109.456 126.141 118.123 253.256 251.446 113.757 249.261 105.029 252.644 251.053 259.184 246.419 248.774 243.462 245.101 242.791 98.962 261.755 111.58 120.149 132.12 208.651 112.539 137.891 132.095 114.937 141.479 116.232 125.959 133.367 104.641 135.882 133.748 109.717 99.486 253.805 243.069 243.966 255.109 182.603 248.388 250.838 253.632 249.401 0 123.563 135.783 188.756 249.613 113.335 132.302 249.139 141.484 123.437 245.595 130.664 251.436 141.212 132.208 96.022 91.142 114.043 105.337 190.167 121.677 142.61 141.857 112.264 0 232.776 236.574 241.497 258.675 206.637 238.143 242.778 189.266 234.247 231.323 89.132 137.251 126.197 250.353 110.645 107.418 127.523 232.776 119.62 117.611 196.504 136.471 256.473 137.464 91.051 124.72 183.439 243.971 106.247 111.193 251.422 122.401 143.151 122.758 255.371 127.199 259.978 126.283 137.179 155.647 111.331 122.948 113.477 106.176 113.563 246.362 256.634 102.355 141.245 245.935 104.211 117.185 105.617 263.674 116.789 246.322 118.348 137.408 134.785 246.183 116.499 111.494 176.196 109.739 250.612 106.054 138.942 124.019 247.419 110.292 106.247 111.007 238.143 113.335 130.974 134.943 178.445 245.073 93.703 249.844 257.716 251.268 249.98 257.229 248.827 251.881 256.571 249.343 257.576 253.513 256.581 255.608 99.166 116.552 190.097 104.018 115.648 118.918 111.041 130.936 116.547 116.02 184.765 246.021 110.292 112.526 251.893 123.575 116.239 108.121 110.878 2000FordMustang2DGT 260.2 189.51 179.686 181.983 185.922 124.29 98.962 182.115 247.805 124.019 129.397 247.388 139.576 131.614 104.258 129.828 199.026 108.587 127.001 121.663 114.739 135.314 118.418 116.997 122.203 109.221 132.231 122.129 110.332 135.504 0 238.386 127.523 128.096 135.502 173.241 235.974 111.007 234.22 238.386 130.37 69.287 253.189 246.443 243.733 255.142 184.153 246.278 96.022 193.412 256.365 138.942 137.892 253.937 0 248.686 247.067 237.736 204.352 250.353 252.932 261.755 181.245 90.468 106.26 122.6 124.025 114.793 125.836 85.26 242.253 135.152 118.119 121.904 139.637 194.782 132.351 131.144 122.959 139.879 145.495 133.592 129.673 138.871 131.501 143.672 130.305 137.951 142.665 89.132 126.919 137.464 184.418 246.107 106.054 128.888 246.602 0 249.515 105.013 110.792 119.436 235.053 126.197 103.854 92.372 109.92 248.623 249.72 253.802 256.473 190.724 106.166 250.612 244.409 242.06 137.251 110.545 254.98 253.958 71.019 178.257 176.196 164.587 188.225 191.263 180.523 182.978 194.814 239.203 189.491 189.964 185.965 192.712 197.806 190.977 190.303 194.555 189.539 196.753 190.915 194.904 196.225 87.677 240.793 128.669 135.197 136.471 171.218 239.053 109.739 0 102.576 259.533 119.407 129.368 133.164 240.59 104.419 247.562 259.533 249.515 115.78 119.62 181.983 242.748 116.499 250.34 252.677 79.566 110.712 234.904 107.575 107.192 117.611 185.922 246.758 111.494 118.231 252.249 121.476 119.451 114.637 112.507 189.368 117.074 112.052 110.887 118.655 124.667 127.942 103.874 111.959 115.992 2000FordFocus4Dwag 82.868 114.508 244.21 250.474 254.233 246.139 255.241 249.801 98.952 102.934 126.967 108.729 125.527 107.067 105.984 122.138 102.782 104.451 123.734 90.468 100.245 0 167.569 236.68 117.929 103.994 137.97 264.21 267.976 262.383 268.768 264.962 269.766 265.497 150.47 132.326 133.727 142.515 128.596 143.823 134.456 131.687 145.036 66.089 253.757 243.202 242.033 254.605 183.506 245.696 250.051 250.719 246.346 254.599 196.55 121.669 258.16 69.019 2000FordFocus4DLX 144.97 273.591 140.382 255.242 137.536 78.318 242.719 140.966 131.535 120.786 141.959 195.943 130.056 134.866 130.783 130.351 114.45 117.258 143.07 137.072 126.805 130.514 132.838 137.309 125.153 136.117 257.82 267.242 92.372 83.879 247.424 101.739 252.684 84.27 244.965 254.069 191.99 119.854 111.167 120.119 120.128 123.244 128.932 108.278 116.975 122.697 117.643 119.875 129.61 187.107 133.987 124.838 118.228 136.251 95.798 246.078 257.432 245.094 86.289 237.721 247.195 124.723 141.152 235.991 94.613 244.925 123.458 110.802 125.755 189.51 251.268 134.785 134.721 250.173 130.477 134.506 110.533 123.477 196.366 110.064 127.853 126.473 102.236 132.779 119.162 120.512 120.939 105.979 2000FordFocus2DZX3 96.834 242.81 253.598 180.389 247.328 250.064 252.315 248.211 257.215 248.541 252.673 256.451 248.974 256.814 252.207 256.925 254.573 252.82 124.483 119.475 119.262 110.377 121.98 252.565 121.745 121.362 107.142 249.584 93.343 134.082 114.165 250.063 117.929 107.575 188.247 128.669 184.37 172.988 174.157 172.956 180.198 191.438 182.664 196.759 180.221 188.021 188.368 172.544 201.343 175.969 183.432 191.959 193.003 93.244 103.68 251.729 243.968 133.11 255.056 118.983 113.065 133.922 107.402 202.382 136.531 106.992 104.795 144.338 107.338 143.352 114.359 112.089 137.062 111.552 117.098 130.427 107.367 93.703 246.322 241.306 191.44 190.196 195.696 194.889 191.989 207.868 180.972 158.87 105.089 110.827 254.67 95.405 252.75 248.957 256.576 248.365 253.584 257.016 249.648 82.868 193.003 257.716 137.408 139.531 253.125 138.951 136.726 114.526 132.605 206.244 115.352 134.356 127.896 113.985 138.632 118.607 120.104 128.801 107.319 133.012 131.101 103.019 138.149 108.19 151.257 268.258 156.218 268.379 145.585 46.894 252.227 246.989 235.817 108.503 236.993 255.34 92.815 113.907 103.039 188.226 121.717 93.343 121.974 137.536 191.959 249.844 118.348 139.581 252.819 100.392 103.697 120.486 102.798 83.807 100.162 259.528 115.325 125.043 133.018 240.574 134.082 103.647 180.92 239.709 109.738 246.022 256.186 248.108 121.98 167.569 76.544 244.149 132.632 138.271 140.382 175.969 236.993 116.789 95.41 139.858 131.083 250.296 113.234 114.666 134.097 234.207 254.31 135.883 97.054 178.602 111.666 248.206 123.039 116.305 0 182.749 115.26 113.026 115.76 128.863 119.397 132.567 121.766 126.692 135.003 258.01 252.189 258.149 91.873 247.47 250.389 131.34 108.974 143.059 198.794 183.724 57.029 107.726 246.048 257.442 253.55 242.512 243.536 255.001 182.543 240.75 129.982 112.834 101.005 196.759 257.442 141.245 141.152 254.069 137.231 143.699 123.538 137.649 206.591 123.098 139.664 130.383 119.556 144.085 126.832 126.922 127.681 121.997 135.858 2000FordEscort4DSedan 254.6 266.396 241.004 244.276 240.628 238.155 236.182 136.79 126.276 142.173 150.106 136.539 133.772 142.092 134.012 147.548 134.135 137.509 147.934 90.023 129.656 107.246 120.55 111.106 134.062 122.641 131.506 252.011 257.427 126.552 112.834 251.254 130.969 120.305 135.265 266.008 138.271 250.832 121.974 103.647 108.472 251.167 103.994 107.192 190.578 135.197 253.802 126.919 110.545 103.854 252.932 105.268 111.678 128.096 236.574 123.563 124.76 128.715 117.899 135.975 131.222 141.375 260.733 263.675 135.496 101.005 261.499 136.092 132.323 134.06 84.27 252.555 245.697 248.713 254.834 184.347 250.063 252.463 255.853 248.194 257.205 246.729 254.148 255.872 251.846 255.888 252.962 261.667 89.636 126.552 135.496 182.664 246.048 102.355 124.723 244.965 2000FordCrownVictoria4D 97.768 98.423 145.44 198.754 108.94 110.488 128.956 184.521 119.575 122.493 118.477 128.289 138.319 122.878 89.416 125.749 109.328 119.084 105.744 102.362 120.878 105.103 52.756 246.064 244.459 235.783 206.539 98.322 247.646 120.285 142.049 131.237 238.632 112.068 104.156 250.38 186.213 240.735 76.544 238.753 134.097 133.018 125.847 237.591 100.245 110.712 167.249 248.72 116.988 105.397 120.105 262.118 132.632 248.709 115.44 131.506 141.375 172.956 236.182 113.563 86.645 118.279 234.763 105.397 120.305 132.323 188.368 246.989 117.185 122.952 63.39 246.078 249.584 180.887 240.75 206.539 232.013 234.763 241.823 217.518 244.149 192.131 234.207 240.574 236.233 208.367 89.636 129.982 133.33 131.297 105.576 122.855 193.982 107.655 122.669 124.086 106.241 133.004 121.398 114.291 123.283 108.771 126.692 123.792 110.508 129.795 93.244 243.462 140.309 125.731 123.934 97.877 109.16 131.115 232.013 116.988 130.969 136.092 188.021 252.227 104.211 180.92 178.602 182.749 99.03 248.108 87.464 240.31 184.37 244.276 122.948 110.827 248.774 89.081 238.597 114.482 122.641 131.222 174.157 238.155 106.176 239.31 88.679 83.807 115.662 256.186 116.305 121.362 120.84 98.59 254.828 120.55 128.715 180.972 241.004 111.331 105.089 246.419 125.431 117.111 107.195 127.206 186.813 115.397 121.793 122.418 118.334 138.412 118.825 117.285 128.343 115.016 129.514 122.008 120.475 134.051 237.54 121.504 111.106 117.899 114.37 254.31 248.334 109.738 248.206 252.565 94.961 121.02 137.549 235.783 131.115 118.279 240.31 115.166 259.49 115.805 144.152 124.133 252.644 0 244.644 248.455 112.22 105.623 138.436 114.913 124.57 131.254 111.632 137.999 131.724 110.593 143.755 130.05 143.441 142.877 131.203 143.962 129.589 130.835 141.008 125.729 141.344 142.322 129.603 144.485 98.903 248.03 113.198 134.618 117.532 95.41 135.883 132.452 246.022 123.039 121.745 185.406 124.154 248.889 95.621 120.458 82.491 237.559 246.852 193.55 129.768 101.555 108.446 129.381 117.179 127.275 112.737 109.823 133.779 158.87 259.184 148.023 148.277 132.239 139.967 215.788 90.601 0 252.582 103.025 108.573 191.566 137.153 248.334 132.452 115.662 250.38 116.097 117.255 108.573 248.455 98.61 258.077 144.152 135.766 142.049 270.812 134.481 252.921 139.858 54.542 107.539 247.457 260.054 97.075 245.769 95.357 240.735 137.297 136.187 137.153 239.709 111.666 259.49 254.099 247.646 121.911 242.886 95.88 94.961 103.329 124.76 207.868 266.396 155.647 82.862 186.241 127.662 254.099 125.475 135.766 117.415 248.256 0 111.496 257.419 115.344 117.255 97.075 250.398 257.419 252.582 74.022 142.768 246.249 115.805 125.475 120.285 259.845 134.861 246.215 95.621 98.512 105.521 131.138 107.819 254.6 137.179 132.208 251.053 140.734 138.868 112.934 135.753 205.675 115.864 136.978 129.094 114.761 143.779 116.341 129.59 87.464 241.778 134.642 134.062 135.975 172.988 240.628 113.477 0 135.686 125.918 250.398 123.098 116.097 192.835 137.297 254.313 252.44 135.686 97.169 119.945 117.532 127.119 248.699 256.082 127.837 109.502 249.058 124.133 117.415 131.237 264.925 125.304 244.445 131.083 100.162 256.64 104.243 135.858 147.186 115.166 121.504 134.642 114.482 259.37 147.359 261.516 135.778 114.574 98.322 257.669 116.7 243.047 101.628 104.283 97.169 246.712 97.768 191.989 242.75 147.186 261.48 271.354 263.741 107.298 261.017 264.954 203.953 257.669 121.911 259.845 270.812 264.925 100.894 263.728 262.256 254.828 217.518 262.118 266.008 273.591 201.343 108.503 263.674 254.137 122.205 263.481 254.774 258.431 265.075 198.481 259.392 263.196 265.924 0 251.733 261.48 143.387 251.733 141.16 147.638 154.045 120.84 114.23 186.673 120.592 0 240.197 143.387 137.773 129.653 239.085 107.077 114.336 167.608 97.096 259.224 138.943 132.005 142.188 271.354 137.773 113.85 130.771 267.705 58.859 243.567 247.119 178.972 240.064 105.698 246.249 258.077 249.058 96.584 243.432 129.097 142.188 134.395 237.613 107.885 109.322 166.989 0 256.169 95.88 103.329 123.25 254.452 130.321 258.593 122.685 127.214 147.638 104.283 119.868 143.909 94.387 95.446 250.036 259.224 251.762 134.19 110.315 132.043 134.073 116.462 137.533 194.59 136.232 121.666 122.998 135.307 135.795 137.754 126.436 137.555 132.187 130.283 131.215 143.599 125.878 260.64 190.601 253.497 256.946 259.452 254.441 262.649 254.006 259.622 263.666 256.627 264.453 258.266 264.866 260.986 256.64 261.516 268.379 195.696 101.739 259.978 251.436 105.029 258.295 249.251 251.525 258.306 191.282 253.439 256.318 261.371 251.386 260.925 250.662 258.429 262.547 255.121 261.362 259.834 265.884 258.365 84.64 107.663 257.636 119.559 127.214 130.112 240.249 135.858 114.574 247.53 128.873 113.573 245.89 122.402 118.082 183.582 123.233 245.851 262.22 122.131 250.701 110.082 132.005 111.298 248.886 105.311 0 255.166 262.22 255.166 138.6 97.096 108.289 258.902 122.582 115.731 198.128 143.025 254.476 142.768 138.17 127.109 128.061 253.347 254.083 130.136 108.289 251.762 135.651 111.298 134.395 263.741 129.653 248.911 125.918 111.496 96.807 252.716 257.636 267.705 243.047 246.712 241.332 135.917 127.681 246.147 130.337 251.382 141.597 130.112 154.045 132.677 127.827 256.933 131.685 90.982 125.52 250.724 255.569 137.1 130.758 139.241 258.874 260.998 134.775 113.85 129.055 113.573 139.653 134.618 139.877 258.571 260.689 136.062 127.405 2000FordContour4D 0 118.391 239.56 249.802 128.666 137.447 237.119 138.898 122.131 138.6 146.157 119.134 128.873 132.338 113.198 112.643 246.477 250.236 2001FordFocus4dwag 2001FordTaurus4DSES 246.93 235.993 106.995 237.119 96.807 259.787 258.593 251.382 208.178 97.699 148.372 143.145 252.716 128.922 122.685 141.597 235.636 104.243 135.778 145.585 194.889 252.684 126.283 141.212 252.644 59.413 246.468 248.983 180.816 240.213 105.535 246.477 258.571 248.699 0 116.269 124.636 268.487 138.898 253.432 117.194 138.943 135.651 252.409 122.053 246.93 116.269 192.82 209.924 174.657 181.019 169.054 167.822 168.013 180.816 196.382 183.582 198.128 178.972 186.673 186.241 166.989 203.953 167.608 186.213 192.835 100.9 252.308 252.106 259.623 245.627 97.699 0 252.796 117.87 145.377 252.796 256.33 122.512 122.066 90.094 238.857 248.858 115.737 142.232 235.993 124.636 118.391 84.33 248.359 102.878 253.806 255.283 266.377 241.636 245.557 241.458 239.087 238.047 2001FordFocus4DSE 240.4 94.085 102.645 90.333 119.773 126.034 114.536 2001FordFocus4DLX 2001FordMustang2DGT 91.639 112.225 104.552 107.97 245.631 258.576 2000VolvoS804D 138.376 130.888 248.652 134.839 253.409 148.216 129.403 158.842 57.552 125.91 251.969 261.238 253.72 134.327 2000VolvoS704D 2003FordFoucs4Dwag 73.236 250.562 127.31 122.998 121.424 112.99 105.788 130.711 112.546 116.776 248.072 0 258.576 145.377 122.066 142.232 270.424 137.447 96.503 134.775 130.136 249.19 251.332 75.821 248.699 132.338 139.653 119.945 241.332 120.458 94.387 255.569 260.998 254.083 110.627 252.282 257.252 196.382 250.831 244.88 83.879 255.371 245.595 113.757 259.223 2000VolvoS404D 89.901 256.844 268.762 271.133 259.105 260.509 259.367 257.363 252.033 103.202 112.941 259.631 270.424 106.995 268.487 255.43 138.073 258.337 129.207 134.826 151.269 256.33 248.858 112.941 249.802 117.87 122.512 115.737 259.631 128.666 191.44 2001VolvoS804D 252.98 130.211 137.747 161.967 267.088 251.891 145.03 136.622 142.235 259.534 261.427 138.838 82.394 247.388 101.606 252.671 256.361 267.555 240.455 244.623 240.877 237.795 235.282 107.97 261.238 0 138.838 245.631 108.19 198.794 257.432 143.151 141.484 253.256 140.154 139.568 114.853 131.257 203.158 109.743 132.665 134.759 108.281 137.478 115.765 119.794 259.37 268.258 2001VolvoS604D 148.084 117.238 248.067 126.115 2003FordFocusZX3 125.06 128.507 115.805 0 248.528 261.427 129.68 113.268 110.464 247.102 248.528 2001VolvoS404D 2003FordFocus4DZX5 93.225 126.31 101.02 167.822 125.52 139.241 128.061 238.047 103.852 113.555 168.013 90.982 250.724 258.874 253.347 2002VolvoS804D 130.956 130.778 99.249 133.915 146.022 112.391 239.56 2002VolvoS604D 148.632 128.734 261.152 145.601 264.743 128.868 145.553 155.109 132.698 2003FordFocus4DZTS 87.886 247.506 259.374 262.564 252.077 252.733 252.309 250.919 243.518 107.814 57.552 251.969 248.072 238.857 103.202 2003VolvoS804D 258.794 246.205 2003FordFocus4DSE 0 107.814 247.102 259.534 93.433 241.877 2003VolvoS604D 2003FordFocus2DZX3 238.92 237.736 90.094 252.033 2003VolvoS404D 105.532 152.196 266.302 159.273 266.657 146.715 240.7 125.91 116.776 2004VolvoS804D 2003FordCrownVictoria4D 253.25 257.028 267.539 242.661 246.546 0 237.736 243.518 110.464 142.235 235.282 98.935 242.393 125.311 130.758 127.109 239.087 103.044 2004VolvoS604D 137.265 121.892 253.587 126.634 253.395 90.182 95.894 181.019 115.939 138.17 241.458 114.632 122.939 169.054 83.914 257.363 2004VolvoS404D 257.127 249.718 112.063 247.175 2004FordTaurus4DSE 98.057 97.754 77.909 2005VolvoS804D 2004FordMustang2D 78.641 247.138 97.999 108.058 238.92 250.919 113.268 136.622 237.795 121.424 112.546 137.1 2005VolvoS604D 259.347 90.182 142.51 2005VolvoS404D 2004FordFoucs2DZX3 0 96.471 242.243 2000FordTaurus4DSE 139.138 116.893 244.634 123.531 248.392 125.626 136.456 155.768 94.552 100.9 253.797 264.963 260.258 192.82 136.715 252.106 126.55 126.034 120.282 241.636 101.793 100.442 174.657 108.293 245.627 119.134 129.055 127.31 105.788 112.225 260.509 102.645 245.835 129.347 114.536 102.844 245.557 145.03 240.877 122.998 130.711 104.552 259.367 2000FordMustang2DGT 94.085 243.888 2004FordFocus4Dwag 77.995 107.954 129.68 2000FordFocus4DLX 91.639 259.105 131.19 130.625 156.079 126.31 115.805 244.623 2000FordFocus4Dwag 112.99 137.749 124.156 249.346 124.501 254.607 240.7 252.309 200.62 254.776 134.36 266.377 138.053 144.701 209.924 159.707 259.623 146.157 2004FordFocus4DZX5 97.754 2000FordFocus2DZX3 97.999 242.661 252.077 112.391 128.507 240.455 132.698 0 118.076 107.954 108.058 246.546 252.733 2000FordEscort4DSedan 90.333 118.731 255.283 119.804 120.872 125.06 267.555 155.109 151.269 161.967 271.133 158.842 259.786 141.583 119.773 2000FordContour4D 93.225 256.361 145.553 134.826 137.747 268.762 129.403 251.303 140.199 240.4 132.677 119.868 123.25 127.681 242.134 127.827 143.909 151.257 183.724 245.094 122.758 123.437 251.446 125.609 108.818 131.146 122.848 78.302 255.575 254.452 246.147 213.063 256.933 138.6 190.509 133.968 249.333 125.365 155.231 146.427 247.309 132.305 130.321 130.337 244.379 131.685 147.359 156.218 190.196 247.424 127.199 130.664 249.261 137.202 115.675 131.953 136.571 197.758 134.602 129.588 135.124 135.967 144.006 131.602 129.777 143.532 137.607 143.188 133.489 149.439 139.069 89.901 253.409 108.027 256.423 265.434 257.911 102.878 256.645 260.929 120.04 132.461 253.735 140.311 256.916 137.676 94.552 120.67 127.405 135.917 2000FordCrownVictoria4D 77.995 252.98 99.249 146.715 252.671 128.868 129.207 130.211 256.844 148.216 254.113 103.991 145.884 137.572 253.806 133.528 129.964 197.209 142.713 252.308 147.145 134.025 251.215 140.143 256.665 148.755 137.725 162.751 110.743 118.076 0 145.92 252.749 138.034 105.877 113.581 259.965 84.33 253.573 257.362 193.622 249.754 115.616 253.084 265.211 257.541 2001FordTaurus4DSES 257.07 266.257 259.742 2001FordFocus4dwag 73.236 248.652 109.297 2001FordMustang2DGT 255.43 248.067 87.886 253.395 266.657 101.606 264.743 258.337 253.25 247.506 113.84 137.725 130.625 136.456 257.028 259.374 133.915 99.741 110.743 2001FordFocus4DLX 0 2001FordFocus4DSE 82.394 261.152 2004FordFocus4DZTW 99.741 2001FordFocus2DZX3 98.057 0 149.414 139.801 162.751 156.079 155.768 267.539 262.564 146.022 113.84 139.801 2001FordEscort4DSedan 267.02 140.889 121.901 2002FordTaurus4DSES 0 121.901 123.083 131.19 125.626 2001FordCrownVictoria4D 267.02 255.631 256.916 256.665 254.607 248.392 0 145.113 140.889 133.758 137.676 148.755 2002FordFocus4Dwag 0 251.392 263.617 2002FordMustang2DGT 78.641 112.063 253.587 266.302 0 246.951 121.102 152.782 150.697 125.111 140.311 140.143 124.501 123.531 247.138 247.175 126.634 159.273 247.388 145.601 138.073 126.115 250.562 134.839 249.126 131.378 154.208 146.017 248.359 136.145 93.01 246.951 2002FordFocus4DZX5 2002FordFocus4DSE 6 120.559 123.436 198.039 245.96 249.718 121.892 152.196 246.205 128.734 130.778 117.238 251.891 130.888 250.633 126.965 145.769 137.942 245.807 125.648 132.637 185.814 126.825 253.154 120.629 145.624 142.068 246.122 127.234 2004FordFocus4DZTS 245.96 2002FordFocus4DZTS 2002FordFocus2DZX3 2002FordEscort4DSedan 2003FordTaurus4dSES 129.717 131.365 251.936 125.111 255.631 133.758 123.083 149.414 2002FordCrownVictoria4D 2004FordFocus4DSVT 2003FordMustang2DGT 110.21 150.339 269.735 150.697 2003FordFocusZX3 2004FordCrownVictoriaSedan4D 2003FordFoucs4Dwag 98.507 147.246 263.499 152.782 263.617 145.113 2003FordFocus4DZX5 2004FordCrownVictoria4DLX 2003FordFocus4DSE 135.582 118.935 254.588 121.102 251.392 2003FordFocus4DZTS 262.195 247.829 2005FordTaurusSE4D 2003FordFocus2DZX3 144.88 120.288 248.132 2005FordMustangCoupe2DV6Deluxe 2003FordCrownVictoria4D 93.01 254.588 263.499 269.735 251.936 253.735 251.215 249.346 244.634 2005FordFocusZX54D 2004FordMustang2D 120.04 147.145 137.749 139.138 259.347 257.127 137.265 105.532 258.794 148.632 130.956 148.084 267.088 138.376 254.185 135.555 103.196 116.005 0 247.557 120.288 247.829 118.935 147.246 150.339 131.365 132.461 134.025 124.156 116.893 0 248.132 2004FordTaurus4DSE 2004FordFoucs2DZX3 2004FordFocus4Dwag 2004FordFocus4DZTS 2004FordFocus4DZX5 110.21 129.717 2004FordFocus4DZTW 98.507 2004FordFocus4DSVT 144.88 262.195 135.582 2004FordCrownVictoriaSedan4D 263.898 247.557 2005FordTaurusSE4D 141.692 2005FordFocusZX32D 2004FordCrownVictoria4DLX 2005FordFocusZX54D 141.692 263.898 2005FordFiveHundred4DSE 2005FordMustangCoupe2DV6Deluxe 2005FordFocusZX32D 2005FordCrownVictoria4D 0 2005FordFiveHundred4DSE 2005FordCrownVictoria4D 93.74 106.33 103.46 84.782 117.325 105.762 0 119.257 121.775 117.859 122.342 93.74 119.257 0 104.362 121.142 81.259 84.782 121.775 104.362 0 122.23 111.792 106.33 117.325 117.859 121.142 122.23 0 128.111 103.46 105.762 122.342 81.259 111.792 128.111 55 0 Preliminary Results Unknown:2005FordFocusZX32D vs Only 2D Vehicles 120 100 80 Series1 60 40 20 0 2002FordMustang2DGT 2000FordMustang2DGT 2004FordMustang2D 2003FordMustang2DGT 2002ToyotaEcho2D 2000ToyotaEcho2D 2002ToyotaCelica2D 2001ToyotaCelica2D 2003ToyotaCelica2D 2000ToyotaCelica2D 2004ToyotaCelica2D 2005FordFocusZX32D 2001FordFocus2DZX3 2004FordFoucs2DZX3 2001ToyotaSolara2D 2003FordFocus2DZX3 2000FordFocus2DZX3 2000ToyotaSolara2D 2002FordFocus2DZX3 2003TotoyaSolara2D 2004ToyotaSolara2D 2005ToyotaSolara2D 2005ToyotaCelica2D 2005FordMustangCoupe2DV6Deluxe 56 Next Steps Preliminary Conclusion: Creating a feature vector composed of shape descriptors using shape contexts demonstrates a high discriminatory power in vehicle identification. ELATION! 57 Question #2 for the Class • What were we to commit to memory as part of class question #1? • A novel application of existing methods does not a dissertation topic make 58 Dissertation Timeline Carl E. Abrams Elation Draft Chap 4&5 Now! Draft Chap 3&4 Defense 1/06 1/05 1/04 9/03 Draft Idea Paper Complete Draft Proposal Advisor Selection Complete Dissertation First Paper at Proposal Pace Day Final Manuscript and Paper Committee Formation Final Draft Chap 1-3 Draft Chap 1&2 59 Desperation! • This is not new it is just an application of what is already known!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!! • Now what do I do????? 60 Dissertation Timeline Carl E. Abrams Desperation --- but never give in Draft Chap 4&5 Now! Draft Chap 3&4 Defense 1/06 1/05 1/04 9/03 Draft Idea Paper Complete Draft Proposal Advisor Selection Complete Dissertation First Paper at Proposal Pace Day Final Manuscript and Paper Committee Formation Final Draft Chap 1-3 Draft Chap 1&2 61 Next Steps Preliminary Conclusion: Creating a feature vector composed of shape descriptors using shape contexts demonstrates a high discriminatory power in vehicle identification. Research Exploration: How well can shape contexts perform in the role of image segmentation??? 1.Develop semantic extraction using Shape Contexts 2.Test segmentation / extraction method for semantic shapes 3.Compare segmentation / extraction method for semantic shapes to other methods 62 Shape Extraction using Shape Contexts 1. 2. 3. 4. 5. From known database develop invariant models for window shape 1,2 and 3, door shapes and body shape using clustering analysis For these known shapes and their contexts, develop shape histograms for neighborhood of high information context points on shape using entropy measures Scan across an edge detected image using a window, computing the local neighborhood in the window Compare the local neighborhood to the known shape contexts local neighborhood and find best fits Focus on best fits and match known shape to points in the best fit area 63 For known shapes and their contexts, develop shape histograms for neighborhood of high information context points on shape using entropy measures • Use entropy calculation to determine the local set of points to use as reference on known shape context. (Note: Shape Context is intentionally biased toward close in points) • Use a subset of points on the known shape, slide around the shape calculating entropy select the highest point subset (“The Entropy Strategy for Shape Recognition”, D Geman, Proc. 1994 IEEE-IMS workshop on Information Theory and Statistics, Alexandria, VA, October, 1994 64 Shape Entropy Example 16 discrete samples from the curve •12 flat line segments – 0 deg turning angle •4 corners - 90 deg turning angle Prob 0 deg =3/4 and Prob 90 deg = ¼ H ( x) pi log pi .8 i Slide window (“The Entropy Strategy for Shape Recognition”, D Geman, Proc. 1994 IEEE-IMS workshop on Information Theory and Statistics, Alexandria, VA, October, 1994 65 Segmentation using Local Shape Contexts by looking for similar local shape histograms Select the closet match and then fit the invariant model to the available points in the image by scaling for best fit 66 Dissertation Timeline Carl E. Abrams Despair -- What cap and gown? Draft Chap 4&5 Now! Draft Chap 3&4 Defense 1/06 1/05 1/04 9/03 Draft Idea Paper Complete Draft Proposal Advisor Selection Complete Dissertation First Paper at Proposal Pace Day Final Manuscript and Paper Committee Formation Final Draft Chap 1-3 Draft Chap 1&2 67 Beyond Desperation Trying to regain the will to live And then: ……. 68 Introduction and Overview • The research focuses on and extends the work done on a new shape descriptor called “Shape Contexts” • Constraining shapes to be continuous outlines (no holes) it is proved using graph theory and then confirmed through experiments that the original matching method which is modeled on the “Assignment Problem” is subject to degenerate shape description • A simpler matching algorithm called the “Least Cost Diagonal” which utilizes the physical relationship of points on a boundary is introduced and compared to the original method • The efficacy of the Least Cost Diagonal method is confirmed through experiments to the original method in a Nearest Neighbor comparison using a previously classified pottery database 69 Shape Contexts-Assumptions • Comparing two shapes was modeled after “The Assignment Problem” • Assumption: The boundary points are in no particular order[2] X X X [2] T. B. Sebastian, P. N. Klein, and B. B. Kimia, "Recognition of shapes by editing their shock graphs," Pattern Analysis and Machine Intelligence, IEEE Transactions on, vol. 26, pp. 550-571, 2004. 70 Shape Contexts-The Assignment Problem • • What is the best fit (minimum cost) to align all the points? The Assignment Problem – Hungarian Method for bi-partite matching problem We will be working with the following problem: assign n = 9 candidates to n=9 jobs to minimize the total salary cost paid by the department. The individual salaries of each candidate at each job position depend on their qualification and are given by the cost matrix (in $ per hour): Sam Jill John Liz Ann Lois Pete Alex Herb Administrator 20 15 10 10 17 23 25 5 15 Secretary to the Chair 10 10 12 15 9 7 8 7 8 Undergraduate Secretary 12 9 9 10 10 5 7 13 9 Graduate Secretary 13 14 10 15 15 5 8 20 10 Financial Clerk 12 13 10 15 14 5 9 20 10 Secretary 15 14 15 16 15 5 10 20 10 Web designer 7 9 12 12 7 6 7 15 12 Receptionist 5 6 8 8 5 4 5 10 7 Typist 5 6 8 8 5 4 5 10 7 If we start with the position of Administrator and assign it to Alex (he gets the minimal salary for this position), then we assign the position of Secretary to Chair to Lois (he gets the minimal salary for this position), and so on, up to the position of Typist, then the assignment is given by the assignment matrix: 71 Shape Contexts-Observations • Using the assignment problem model means that the neighborhood-ness of the boundary points is not used. • The potential of degenerate shape description exists ( i.e. very different shapes that use the same boundary points but in which the points occurs in different orders will be computed to be the same) 72 Shape Contexts-Degenerate Behavior X X 3 3 X X 4 2 X 5 X X 7 6 X X X X X 1 X 5 4 2 3 X X X 7 6 X X X 7 6 4 2 X 1 X 1 Point X Y Point X Y Point X Y 1 0.00 0.00 1 0.00 0.00 1 0.00 0.00 2 -.7071 .7071 2 -.7071 .7071 2 -.7071 .7071 3 0.00 1.4142 3 0.00 1.4142 3 .866 .5 4 .866 .9142 4 .866 .5 4 0 1.4142 5 1.866 .9142 5 .866 .9142 5 .866 .9142 6 1.866 .500 6 1.866 .9142 6 1.866 .5 7 .866 .500 7 1.866 .500 7 .866 .500 Shape 1 5 Shape 2 Shape 3 Distance from Shape 1 – Hungarian 0 Distance from Shape 1 – Hungarian 0 Distance from Shape 1 – LCD 25 rotation 0 degrees Distance from Shape 1 – LCD 26 rotation 50 degrees 73 Shape Contexts-Degenerate Behavior Formalization Graph Theoretic Proof with Examples Jordan Curve Theorem Step 1: Using the definition of a Hamiltonian Cycle prove using If J Jordan's is a simpleTheorem closed curve in each R2, then the Jordan curve theorem, that different Hamiltonian Cyclealso of acalled set the Jordan-Brouwer 1966) states that J Hamiltonian -has two of vertices is atheorem distinct (Spanier shape corresponding toR2 that components Cycle (an "inside" and "outside"), with J the boundary of each. The Jordan curve theorem is a standard result in algebraic topology with a rich history. A complete proof can be found in Hatcher (2002, p. 169), or in classic that the Hamiltonian Cycles of awas particular textsStep such2: asProve Spanier (1966). Recently, a proof checker used bygraph a which have team beentoshown represent distinct shapes aretheorem Japanese-Polish create to a "computer-checked" proof of the (Grabowski 2005). computed to be identical by the shape context method developed by Belongie (solving point assignment matching using the http://mathworld.wolfram.com/JordanCurveTheorem.html Assignment Problem as the model) 74 Shape Contexts-Proof of Degenerate Behavior Step 1 Let a shape S on a binary image I be a connected component where every pixel whose value is 1 is an element of S. A shape S partitions the two dimensional plane into two regions: the interior (union of pixels whose value is 1) and exterior (union of pixels whose value is 0). Lemma 1: Every Hi for a unit graph G forms a certain shape Si . Proof: Hi is a closed cycle graph by definition or an n-gon if G is planar. By the Jordan curve theorem [12,[4]15], Hi partitions the plane into two regions, the interior and exterior. Painting the interior and exterior regions with black and white colors, respectively, produces a connected component shape. Thus, every H i for a unit graph G forms a certain shape Si . [4] O. Veblen, "Theory on plane curves in non-metrical analysis situs," Transactions of the American Mathematical Society vol. 6, pp. 83-98, 1905. 75 Shape Contexts-Proof of Degenerate Behavior Step 1 Example G H1 H2 H3 H4 H5 H6 H7 76 Shape Contexts-Proof of Degenerate Behavior Step 2 Theorem 1: All shapes Si formed by Hi for a unit planar graph G are considered identical shapes by SC algorithm if p1 V . Proof: Let S and S’ be two distinctive shapes formed by two Hamiltonian cycles for a unit graph G where V {v1 vn} . The SC algorithm will place pi s exactly on vertices on G since all edges are unit distance d in length. Let h and h’ be polar histograms for S and S’, respectively. Since locations of vertices are the same in two shapes, h = h’. Hence, we can derive the following from the eqn (2). b Ci ,i 1/ 2 [h(k ) h '(k )] 2 /( h( k ) h '( k )) 0 where i v1 vn k 1 SCdistance ( S , S ') vn C i ,i 0 i v1 Therefore, all shapes Si formed by Hi for a unit planar graph G are considered identical shapes by the SC algorithm if p1 V . ? 77 Shape Contexts-Proof of Degenerate Behavior Step 2 Example S1 S2 S3 S1 S1 (a) S1 S2 S3 S1 S2 S3 0 105 0 105 0 106 0 106 0 (b) 78 The Least Cost Diagonal • Constrains the domain of shapes to outlines (no holes) • Constrains matches to made up of boundary ordered points • Avoids degenerate shape matching • Runs much more quickly than the original method • Gives an indication of degree of rotation of one shape to another 79 The Least Cost Diagonal-How it works x x x x x x x x x x x x x x x Point Point 2 3 4 5 6 1 0.00 6.00 10.00 10.00 10.00 4.00 1 2 6.00 0.00 6.00 6.00 10.00 6.00 2 1 2 3 4 5 6 6.00 4.00 0.00 8.00 3.33 10.00 8.00 6.00 8.00 0.00 6.00 10.00 6.00 10.00 10.00 8.00 6.00 3.33 6.00 0.00 10.00 6.00 6.00 0.00 6.00 6.00 4 6.00 10.00 10.00 10.00 10.00 0.00 5 10.00 10.00 10.00 6.00 0.00 6.00 5 0.00 8.00 6.00 8.00 8.00 6.00 6 6.00 10.00 6.00 6.00 0.00 6 8.00 0.00 4.00 6.00 6.00 10.00 0 degrees 90 degrees x x x x x x x x x x x x x x x x x x x x 4.00 3 x 6.00 Point Point 0.00 4 10.00 3 10.00 x 1 x x x x x x x x x x x Point Point 1 1 2 3 4 5 6 6.00 4.00 0.00 8.00 3.33 10.00 6.00 8.00 0.00 8.00 6.00 10.00 3 8.00 6.00 3.33 6.00 0.00 10.00 4 6.00 10.00 10.00 10.00 10.00 0.00 Point Point 2 1 2 3 4 5 1 3.33 10.00 6.00 4.00 0.00 8.00 2 6.00 10.00 8.00 6.00 8.00 0.00 3 0.00 10.00 8.00 6.00 3.33 4 10.00 0.00 6.00 10.00 10.00 8.00 6.00 0.00 8.00 6.00 8.00 6.00 10.00 8.00 0.00 4.00 6.00 5 0.00 8.00 6.00 8.00 8.00 6.00 5 6 8.00 0.00 4.00 6.00 6.00 10.00 6 270 degrees degrees 180 6 180 degrees 270 degrees 6.00 10.00 80 Shape Contexts-Degenerate Behavior X X 3 3 X X 4 2 X 5 X X 7 6 X X X X X 1 X 5 4 2 3 X X X 7 6 X X X 7 6 4 2 X 1 X 1 Point X Y Point X Y Point X Y 1 0.00 0.00 1 0.00 0.00 1 0.00 0.00 2 -.7071 .7071 2 -.7071 .7071 2 -.7071 .7071 3 0.00 1.4142 3 0.00 1.4142 3 .866 .5 4 .866 .9142 4 .866 .5 4 0 1.4142 5 1.866 .9142 5 .866 .9142 5 .866 .9142 6 1.866 .500 6 1.866 .9142 6 1.866 .5 7 .866 .500 7 1.866 .500 7 .866 .500 Shape 1 5 Shape 2 Shape 3 Distance from Shape 1 – Hungarian 0 Distance from Shape 1 – Hungarian 0 Distance from Shape 1 – LCD 25 rotation 0 degrees Distance from Shape 1 – LCD 26 rotation 50 degrees 81 Experimental Results • Basic Shape Context Experiments • Classification Performance • Timing and Boundary Point dependence 82 Experimental Results • Basic Shape Context Experiments – Translation, Scale and Rotation Invariance – Shape Matching with the Least Cost Diagonal – Demonstration of Degenerate Shape Description/Matching 83 Experimental Results • Classification Performance – Use a NN analysis on a known database which had already been grouped into classes[5]. Compare the Assignment Model Matching to the Least Cost Diagonal at varying number of boundary points – Experiments performed: • One pottery class against each other class • One pottery class against groups of two classes • One pottery class against groups of three classes What happened to the cars????? [5] RG. Bishop, S. Cha, and C.C. Tappert, "Identification of Pottery Shapes and Schools Using Image Retrieval Techniques," Proc. MCSCE, CISST, Las Vegas, NV, June 2005. 84 Experimental Results • Timing and Boundary Point dependence – Timing: From all the classification runs plus additional runs at higher number of boundary points verify the order of computational complexity – Point dependence: Run the classification on particular classes that did not perform well at additional boundary points(150 and 250) to demonstrate that a relationship between the number of points and performance exists 85 Classification Experimental Results • Experimental Procedure: Compute % correctly classified for various numbers of boundary points (3,4,5,10,20,100)[6,7] Class 1 Shapes SA1 SA2 SA3 SA4 SA5 Class 2 Shapes SA6 SA7 SB1 SB2 SB3 SB4 SB5 Class 1 Shapes SA1 SA2 SA3 SA4 SA5 SA6 SA7 [6] S. Belongie, J. Malik, and J. Puzicha, "Matching Shapes," presented at International Conference on Computer Vision (ICCV'01) Volume 1 2001. [7] K. Mikolajczyk and C. Schmid, "A performance evaluation of local descriptors," Pattern Analysis and Machine Intelligence, IEEE Transactions on, vol. 27, pp. 1615-1630, 2005. 86 Classification Experimental Results 100 100 90 90 80 80 70 70 60 Diagonal 50 Hungarian 40 % Correct % Correct Example: Class 1 vs Class 2 and Class 3 60 Hungarian 40 30 30 20 20 10 10 0 0 100points 20points 10points 5points 4points 3points 0points 100points 20points 10points 5points 4points 3points 0points Number of Points Diagonal 50 Number of Points 87 Classification Experimental Results: Timing • Averaged the timing for each set of boundary points-60 runs per boundary point. (Added six additional runs to extend the timings up to 250 points) • Fitted a polynomial to the data to derive the computational order of complexity Least Cost Diagonal Timing Hungarian Timing 0.02 Time in Seconds Time in Seconds 250 200 150 100 50 0 -50 0 0.015 0.01 0.005 0 100 200 300 400 Number of Points 500 600 0 100 200 300 400 500 600 Number of Points 88 Classification Experimental Results: Timing Curve Fitting 89 Classification Experimental Results: Boundary Points Class 1 Class 5 90 Classification Experimental Results: Boundary Points Class 1 vs Class 5 250 boundary points 100 100 90 90 80 80 70 70 60 Diagonal 50 Hungarian 40 % Correct % Correct Class 1 vs Class 5 60 250points 150points 100points 20points 10points 5points 4points 3points 0points 100points 20points 10points 5points 0 4points 10 0 3points 20 10 0points 30 20 a Hungarian 40 30 Number of Points Diagonal 50 Number of Points b 91 Classification Experimental Results: Boundary Points Class 3 vs Class 11,12,13 250 points 100 100 90 90 80 80 70 70 60 Diagonal 50 Hungarian 40 % Correct % Correct Class 3 vs Class 11,12,13 60 Hungarian 40 30 30 20 20 10 10 0 0 250points 150points 100points 20points 10points 5points 4points 3points 0points 100points 20points 10points 5points 4points 3points 0points Number of Points Diagonal 50 Number of Points 92 Summary •The research has confirmed the properties of the Shape Context and proved that when the domain of the shape matching is limited to outlines, erroneous shape matching based on degenerate shape descriptors can occur •A new matching method has been introduced( Least Cost Diagonal) and its efficacy verified against real images and compared to the original method developed by Belongie •Further work was identified: •Is there a relationship between the number boundary points to the accuracy of shape matching? •Are there better distance measures for shape context histograms? •Are there better quantizing schemes for shape contexts? 93 Dissertation Timeline Carl E. Abrams Defense was on July 14th, 2006 Draft Chap 4&5 Now! Draft Chap 3&4 Defense 1/06 1/05 1/04 9/03 Draft Idea Paper Complete Draft Proposal Advisor Selection Complete Dissertation First Paper at Proposal Pace Day Final Manuscript and Paper Committee Formation Final Draft Chap 1-3 Draft Chap 1&2 94 Post-Defense Work • Apply the new method to my existing car database – Compare to “Hungarian Method” for accuracy and speed • Make the dissertation longer 95 Discussion/Lessons Learned • • • • • • • Really have an interest in your topic!! Get comfortable with uncertainty See your faculty advisors early and often Work on your project ever day!!! Don’t throw anything away Make it at least 100 pages long without the references “The way to get good ideas is to get lots of ideas, and throw the bad ones away” – Dr. Linus Pauling (The trick is recognizing the bad ones)! Dr. Carl Abrams – DPS Class of 2006) 96 An Overview of IBM Research IBM Research Global Technology Outlook 2013 - Do Not Distribute © 2013 IBM Corporation GTO 2013 Introduction IBM Research Overview 60 Years of Innovation 1970s 1980s Corporate funded research agenda Technology transfer 1990s Collaborative team Shared agenda Work on client problems Effectiveness 2000s Create business advantage for clients Industry-focused research Research Partnership Innovation Labs Research Institutes Collaborative Innovation Research in the Marketplace Joint Programs Research Services Centrally Funded Collaborate on client-specific technology and business solutions Global Technology Outlook 2013 - Do Not Distribute © 2013 IBM Corporation FOAK First of a Kind 98 GTO 2013 Introduction Diversity of Disciplines at IBM Research Behavioral Sciences Materials Science Chemistry Mathematical Sciences Computer Science Physics Electrical Engineering Service Science, Management & Engineering Science & Engineering Technology Innovation Business Innovation Social Innovation Demand Innovation Social & Cognitive Sciences Global Technology Outlook 2013 - Do Not Distribute © 2013 IBM Corporation Business & Management Economics & Markets 99 GTO 2013 Introduction IBM scientists first to distinguish individual molecular bonds Zurich, Switzerland, 14 September 2012—IBM (NYSE: IBM) scientists have been able to differentiate the chemical bonds in individual molecules for the first time using a technique known as non-contact atomic force microscopy (AFM). Global Technology Outlook 2013 - Do Not Distribute © 2013 IBM Corporation 100 GTO 2013 Introduction President Obama Awards IBM Scientists with National Medal of Technology and Innovation for Inventing the Underlying Technology in LASIK Surgery WASHINGTON, D.C. - 01 Feb 2013: President Obama will honor a team of three IBM (NYSE: IBM) scientists -- James J. Wynne, Rangaswamy Srinivasan and Samuel Blum -with the National Medal of Technology and Innovation, the country's most prestigious award given to leading innovators for technological achievement. They are receiving this award for their discovery of a new form of laser surgery, using an excimer laser that made modern LASIK and PRK refractive eye surgery possible. Global Technology Outlook 2013 - Do Not Distribute © 2013 IBM Corporation 101 GTO 2013 Introduction Focus Areas at IBM Research Industry Solutions Cybersecurity Cloud Business Analytics Future Systems Global Technology Outlook 2013 - Do Not Distribute Processors / Storage / Switching © 2013 IBM Corporation Nanotechnolgoy 102 GTO 2013 Introduction Financial Services Industry Solutions Research Banking, Financial Markets, Insurance Platforms Systemic Risk & Regulation LDAR MRAN Industry @ Scale Compliance Accelerator Content Ingestion Liquidity Risk LRX Collateral, Settlement Risk Transactions @Scale Fraud Surveillance Operational & IT Risk quantification & predictive analytics Cyber Security Scalable and Resilient Business Services Complex adaptive systems Client Services Client Lifetime Value Mobile and eCurrency Smarter Commerce ‘Edge’ Analytics Marketing Optimization –Next Best Action SMARC/Social Business Analytics/Simulation/Big Data/Visualization Technology Platforms/Cognitive Computing/Cloud Institutes and Co-laboratory Global Technology Outlook 2013 - Do Not Distribute © 2013 IBM Corporation 103 GTO 2013 Introduction FOAK Projects Banking Financial Markets Insurance 2012 – iPro( Intraday Cash Settlement Predictive Analytics) – Edge Analytics( Real time customer analytics ) 2011 – SMARC Social Media Analytics for Retail Customers 2010 – Advanced Digital Wallet for Financial Services – Social Empathy Selling for Banking 2009 – Systemic Risk Facility – Risk Adjusted Performance Management 2008 – Information Risk and Compliance Monitor – Core Banking Transformation 2007 – VofC (Real time speech analytics in the call center) – Touch-less Secure Desktop – Business of IT Dashboard 2006 – Predictive Application Life Cycle Management 2005 – Operational Risk Modeling and Quantification – SCORE: Symbiotic Context Oriented Information Retrieval 2004 – On Demand Banking transformation Solution – Smart Customer Interaction 2003 – Conversational Biometrics 2012 – MRAN (Multi level Risk Analytic Network) 2011 – Mark to Liquidity (LiRA) 2010 – Loan Data Analytics for Risk (LDAR) 2009 – Systemic Risk Facility 2007 – High Performance Stream Processing for Financial Services – Integrated Power, Cooling and IT Management for Data Centers 2006 – Mirage: Data Center Storage Optimization 2005 – Fossilization of Business Records 2003 – Monte Carlo Grid for Risk Management 2002 – Grid Architecture for Financial Services 1998 – Voice Recognition enabled Customer Relationship Intelligence 2012 – LAQORM: Linking & Analysis of Quantifiable Operations Risk Metrics 2009 – Operation Integrity and Compliance Management 2008 – Continuous Compliance 2007 – Optimizing Business Rules Creation and Validation in Claims Processing 2006 – Information Integration of Distributed Warehouse Information 2005 – Predictive Model Management Middleware – Rule-based Claims Processing 2004 – Insurance Application Transformation Based on Business Rule Extraction and Clustering 2003 – On Demand Business Process integration for Financial Services: Core Insurance 1999 – XML Based Insurance 1998 – Virtual Market Place – Mobile Insurance Worker – Data Hiding for Automobile Insurance Claim Process Source: http://w3.research.ibm.com/FOAKS/foak1.html Global Technology Outlook 2013 - Do Not Distribute © 2013 IBM Corporation 104 GTO 2013 Introduction Computation Power , Storage, Bandwidth Global Technology Outlook 2013 - Do Not Distribute © 2013 IBM Corporation 105 IBM RESEARCH Jeopardy - The IBM Challenge ► 1997 - Chess ■ A finite, mathematically well-defined search space ■ Large but limited number of moves and states ■ Everything explicit, unambiguous mathematical rules ► 2011 - Human Language ■ Ambiguous, contextual and implicit ■ Grounded only in human cognition ■ Seemingly infinite number of ways to express the same meaning [106] 2011-02-23 IBM RESEARCH DeepQA: Massively Parallel Probabilistic Evidence-Based Architecture Question 100s sources 100s Possible Answers 1000’s of Pieces of Evidence 100,000’s scores from many simultaneous Text Analysis Algorithms Multiple Interpretations Question & Topic Analysis Question Decomposition Hypothesis Generation Hypothesis Generation Hypothesis and Evidence Scoring ... [107] Hypothesis and Evidence Scoring Synthesis Final Confidence Merging & Ranking Answer & Confidence 2011-02-23 IBM RESEARCH Art of the Possible: Transforming Healthcare with IBM Watson ► "IBM's work with WellPoint and Memorial Sloan-Kettering Cancer Center represents a landmark collaboration in how technology and evidence-based medicine can transform the way in which healthcare is practiced. These breakthrough capabilities bring forward the first in a series of Watson-based technologies, which exemplifies the value of applying big data and analytics and cognitive computing to tackle the industries most pressing challenges." ► Manoj Saxena, IBM General Manager, Watson Solutions [108] 2011-02-23 IBM RESEARCH The fourth dimension of Big Data: Veracity – handling data in doubt Volume Velocity Variety Veracity* Data at Rest Data in Motion Data in Many Forms Data in Doubt Terabytes to exabytes of existing data to process Streaming data, milliseconds to seconds to respond Structured, unstructured, text, multimedia * Truthfulness, accuracy or precision, correctness [109] Uncertainty due to data inconsistency & incompleteness, ambiguities, latency, deception, model approximations IBM RESEARCH The Challenge of Big Data and its Implications ► Extremely Large Volumes ► Platforms to Process ► Disparate Sources and Forms ► Analyze both Structured and Unstructured Data ► Near Real-Time ► Ingest in Real-Time ► Short Lived Value ► Analytics in Real-Time ► Inherently Ambiguous ► New forms of Analytics [110] S&D/Research SMARC Solution Architecture Public Data Data Ingestion Adapters Twitter Online social media analytic libraries Promotion Scoring Output Data Blogs/Boards SystemT, Midas, SystemML MySpace Google Buzz Online promo stream InfoSphere Streams Facebook Protocols for data & control exchange Customer Models Private Data Transaction Data Offline social media analytic libraries Consumer List Promotion models SystemT, Midas, SystemML Product Categories InfoSphere BigInsights Public consumer profiles mapped to internal Customer ID or Identified as prospects Promotion Information Client Data 111 2011 FOAK Program Existing analytics assets leveraged IBM Confidential FOAK Assets to be developed © 2011 IBM Corporation Email from customer Customer Insights Web store Social media data capture Comments by customer on Facebook, Twitter Customer interact via multiple channels 112 [112] Customers purchase history and other past interactions across channels use to infer general insights about the customer such as lifetime value and tastes Next Best Action Decisioning Module Contact Center Agent Customer’s smart phone Customer Real-Time State Store staff’s smartphone From Customer’s Smartphone Email to customer Contact center agent inputs Business rules operating on extracted customer insights and customer state generate NBA recommendations Service Staff Channel Backend Systems POS data Composite view of recent interactions between customer and business captured as “customer state” – such as - immediate goal, sentiment, urgency. Deliver Action to channel Channel Backend Systems Multi-channel Next Best Action (MNBA): Business Problem Instructions to respond on facebook IBM RESEARCH Examples of Actions : Replace broken item (rather than repair), OR offer certain discount on upgrade 2011-02-23 IBM Business Analytics and Optimization MNBA realizes state-based marketing via NBA Optimizer Can be viewed as a path through the stages of shopping Customer is in some state (his/her attributes) at any point in time Enterprise's action will move customer into another state Enterprise's goal is to take sequence of actions to guide customer's path to maximize customer's lifetime value Reinforcement Learning produces optimized targeting rules of the form If customer is in state "s", then take marketing action "a" Customer state “s” represented by current customer attribute vector Campaign E Customer A’s path under… Current marketing policy Optimized marketing policy Potentially Valuable Loyal Customer Loyal Customer Repeater Repeater Repeater One Timer Defector Defector Campaign C Campaign A Bargain Hunter Campaign B © 2012 IBM Corporation Valuable Customer Campaign D 113 IBM RESEARCH Temporal Causal Modeling for Operational Risk Quantification Aggregate Analysis (Process & Business Entity Level) Induced Loss Distribution Advanced Scenario Generation: Probabilistic Graphical Model Structure, Parameters; Elicited from Experts Optimal Mitigation Portfolio Analysis Causal (KRI -> Loss Event) Analysis Capture Risk Mitigation Costs with residuals Common Data Services API Data Sources [114] 2011-02-23 IBM RESEARCH IBM Unveils Cognitive Computing Chips… ARMONK, N.Y., - 18 Aug 2011: Today, IBM (NYSE: IBM) researchers unveiled a new generation of experimental computer chips. IBM’s first neurosynaptic computing chips recreate the phenomena between spiking neurons and synapses in biological systems, such as the brain, through advanced algorithms and silicon circuitry. Its first two prototype chips have already been fabricated and are currently undergoing testing. Called cognitive computers, systems built with these chips won’t be programmed the same way traditional computers are today. Rather, cognitive computers are expected to learn through experiences, find correlations, create hypotheses, and remember – and learn from – the outcomes, mimicking the brains structural and synaptic plasticity. Each core contains 256 clusters of transistors and thousands of random access memory (RAM) elements Computational elements and RAM are wired together RAM emulates neuron, computation emulates synapse [115] 2011-02-23