CAP 5415 Computer Vision. Fall 2011

Transcription

1 CAP 5415 Computer Vision Fall 2011

2 General Instructor: Dr. Mubarak Shah Office: 247-F HEC

3 Course Class Time Tuesdays, Thursdays 12 Noon to 1:15PM 383 ENGR Office hours Tuesdays 1:15 PM to 2:00 PM Thursdays 11 AM to 12 Noon And by appointment Grading Midterm 20% Final 30% Assignments 10% Programs 40% Grading Policy: = A; = B; = C;

4 Course Reference Texts: Mubarak Shah, "Fundamentals of Computer Vision". Richard Szeliski, "Computer Vision: Algorithms and Application, Springer. Emanuele Trucco, Alessandro Verri, "Introductory Techniques for 3-D Computer Vision", Prentice Hall, Course Slides from Previous Years

5 Topics We ll Cover Image Filtering, Edge Detection, Interest Point Detectors Motion and Optical Flow Region Segmentation Object Detection and tracking Line and Curve Detection Shape Analysis Stereopsis Imaging Geometry, Camera Modeling and Calibration We may change order

6 Computer Vision The ability of computers to see. Image Understanding Machine Vision Robot Vision Image Analysis Video Understanding

7 A picture is worth a thousand words.

8 A word is worth a thousand pictures. A HUNT

9 Image 2-D array of numbers (intensity values, gray levels) Gray levels 0 (black) to 255 (white) Color image is 3 2-D arrays of numbers Red Green Blue Resolution (number of rows and columns) 128X X X X480

10

11 Image Formats TIF PGM PBM GIF JPEG

12 Video Sequence of frames 30 frames per second Formats AVI MPEG Quick Time

13 Video Clip

14 Sequence of Images

15 Image Formation Light Source Camera (extrinsic and intrinsic parameters) Scene (Surface reflectance, Surface shape )

16 Perspective Projection (Pin Hole) Image Plane f Lens (X,Y,Z) World point image y Z

17 Orthographic Projection Image Plane (X,Y,Z) World point image y

18 Shape from X Recover 3-D shape from 2-D image(s) Stereo Motion Shading Texture Contours

19 Stereo

20

21 Renault Stereo Pair

22 Depth Map

23 Stereo Pair

24 Shape from Shading

25 Lambertian Model S=L, light source I=S.N

26 Vase (1, 0, 1) (-1, 1, 1) (-1,-1, 1)

27 Shape from Texture

28 Visual Motion

29 Hamburg Taxi seq (Optical Flow)

30 Optical Flow Field Examples

31 Sequence Raw Optical flow 31

32 Video Clip & Mosaic

33 Structure From Motion Reconstructed Shape

34

35 Applications of Computer Vision Face Recognition Object Recognition Video Surveillance and Monitoring Object detection, tracking and behavior analysis Remote Sensing: UAVs Robotics Computer Graphics

36 Face Recognition

37 Object Recognition Finding People in images Problem 1: Given an image I Question: Does I contain an image of a person?

38 Yes Instances

39 No Instances

40 Localize People (Human Detection)

41 Human Detection

42 Airplanes

43 Motor Cycles

44 FACIAL EXPRESSIONS RAISE EYE BROWS SMILE

45 Detecting Driver Alertness

46 Lipreading

47 Video Surveillance and Monitoring Object detection Object tracking Object categorization and classification Event or Activities Recognition Automated Surveillance System (Detection & Tracking)

48 COCOA COCOA System Flow Aerial Video Telemetry* Ego Motion Compensation Feature based + Gradient Based Motion Detection Accumulative Frame Differencing + Background Modeling + Object Segmentation Object Tracking Kernel Tracking + Blob Tracking + Occlusion Handling Registered Images Motion Detection Tracks Event Detection & Indexing

49 Ego Motion Compensation Results - I Aerial Video - EO Mosaic Alignment Mask

50 Ego Motion Compensation Results - II Aerial Video - IR Mosaic Alignment Mask

51 Detection Result

52 Tracking Results

53 Tracking Results

54 UCF YouTube Action Dataset Cycling Diving Golf Swinging Riding Juggling Basketball Shooting Swinging Tennis Swinging Volleyball Spiking Trampoline Jumping Walking Dog

55 Making A Sandwich

56 Human Behavior Recognition

57 Key Frames Sequence 1 (350 frames), Part 1

58 Robot Vision (Unmanned Ground Vehicle) UGV

59 Geo-registration

60 Geo-registration

61 Layer Based Video Composition

62 Results of Doll

63 Results of Mom-Daughter