Successfully reported this slideshow.
We use your LinkedIn profile and activity data to personalize ads and to show you more relevant ads. You can change your ad preferences anytime.

General introduction to computer vision


Published on

  • Login to see the comments

General introduction to computer vision

  1. 1. CS 591 E / CS 791 L (CRN: 18390 / 18490) Computer Vision Instructor: Guodong Guo [email_address]
  2. 2. Welcome! <ul><li>Introductions </li></ul><ul><li>Administrative Matters </li></ul><ul><li>Course Outline </li></ul><ul><li>Applications of Computer Vision </li></ul><ul><li>Computer Vision Focus </li></ul><ul><li>Computer Vision Publications </li></ul><ul><ul><li>Journals </li></ul></ul><ul><ul><li>Conferences </li></ul></ul>
  3. 3. Instructor <ul><li>Guodong Guo </li></ul><ul><li>Ph.D. in CS from UW-Madison </li></ul><ul><li> </li></ul><ul><li>Major Research Interest </li></ul><ul><ul><li>Computer Vision, Machine Learning, Pattern Recognition, Biometrics, Multimedia, and HCI </li></ul></ul>
  4. 4. About You … <ul><li>What do you know already? </li></ul><ul><ul><li>C/C++ (Visual C++) </li></ul></ul><ul><ul><li>Matlab </li></ul></ul><ul><ul><li>Images </li></ul></ul><ul><ul><li>OpenCV </li></ul></ul><ul><ul><li> </li></ul></ul><ul><ul><li>Install OpenCV in your PC or laptop, </li></ul></ul><ul><ul><li>Read the manual introduction </li></ul></ul><ul><ul><li>Try to load and save images (homework #0) </li></ul></ul>
  5. 5. Outline <ul><li>Introductions </li></ul><ul><li>Administrative Matters </li></ul><ul><li>Course Outline </li></ul><ul><li>Applications of Computer Vision </li></ul><ul><li>Computer Vision Focus </li></ul><ul><li>Computer Vision Publications </li></ul>
  6. 6. Meeting Times <ul><li>Lectures </li></ul><ul><ul><li>M 17:00-19:30 pm </li></ul></ul><ul><ul><li>Room ESB-E 449 </li></ul></ul><ul><li>Office hours </li></ul><ul><ul><li>TR 1:00-2:00 pm (ESB 753)? </li></ul></ul><ul><ul><li>Or by appointment </li></ul></ul>
  7. 7. Grading <ul><li>The final grade depends on: </li></ul><ul><ul><li>Homework and programming assignments: 40% </li></ul></ul><ul><ul><li>Exams (Midterm): 40% </li></ul></ul><ul><ul><li>Final project (may include class presentation): 20% </li></ul></ul><ul><ul><li>Class participation: (-5%, if absent >= 3times) </li></ul></ul><ul><ul><li>Extra: 1~10% (for creative ideas, paper submission, etc.) </li></ul></ul>
  8. 8. Textbook <ul><li>Computer Vision: A Modern Approach , 2 th Edition, by David Forsyth and Jean Ponce, Prentice Hall, 2003 </li></ul>
  9. 9. Look at the Syllabus <ul><li>Course Objectives </li></ul><ul><li>Expected learning outcomes </li></ul><ul><li>Detailed list of topics (maybe updated) </li></ul>
  10. 10. Outline <ul><li>Introductions </li></ul><ul><li>Administrative Matters </li></ul><ul><li>Course Outline </li></ul><ul><li>Applications of Computer Vision </li></ul><ul><li>Computer Vision Focus </li></ul><ul><li>Computer Vision Publications </li></ul>
  11. 11. What is Computer Vision? <ul><li>Given an image or more, extract properties of the 3D world </li></ul><ul><li>Traffic scene </li></ul><ul><li>Number of vehicles </li></ul><ul><li>Type of vehicles </li></ul><ul><li>Location of closest obstacle </li></ul><ul><li>Assessment of congestion </li></ul>
  12. 12. Computer Vision vs. Graphics <ul><li>3D  2D implies information loss </li></ul><ul><li>sensitivity to errors </li></ul><ul><li>need for models </li></ul>graphics vision
  13. 13. Computer Vision vs. Biometrics <ul><li>Biometrics comprises methods for uniquely recognizing humans based upon one or more intrinsic physical or behavioral traits </li></ul><ul><ul><li>Physiological are related to the shape of the body, e.g., fingerprint, face recognition, DNA, hand and palm geometry, iris recognition, which has largely replaced retina, and odor/scent </li></ul></ul><ul><ul><li>Behavioral are related to the behavior of a person, e.g., typing rhythm, gait, and voice </li></ul></ul>
  14. 14. Computer Vision vs. Biometrics <ul><li>Biometrics is a branch of Computer Vision </li></ul><ul><li>The development of Biometrics depends on Computer Vision techniques </li></ul>
  15. 15. Computer Vision vs. Machine Learning <ul><li>Machine learning is a scientific discipline that is concerned with the design and development of algorithms that allow computers to change behavior based on data, such as from sensor data or databases (from Wikipedia) </li></ul><ul><li>A major focus of machine learning research is to automatically learn to recognize complex patterns and make intelligent decisions based on data. </li></ul>
  16. 16. Computer Vision vs. Machine Learning <ul><li>Machine Learning is very useful for Computer Vision (e.g., learning for vision) </li></ul><ul><li>Computer Vision is more than just learning </li></ul><ul><ul><li>Modeling </li></ul></ul><ul><ul><li>Example based learning </li></ul></ul><ul><li>In Machine Learning, it usually does not care about how to obtain the data or sensors </li></ul><ul><li>In Computer Vision, we care how to obtain the visual data (sensor design, active vision), how to represent the visual data, and others </li></ul>
  17. 17. Vision <ul><li>Vision is the process of discovering what is present in the world and where it is by looking. </li></ul>
  18. 18. Computer Vision <ul><li>Computer Vision is the study of analysis of pictures and videos in order to achieve results similar to those as by people. </li></ul>
  19. 19. Why Computer Vision <ul><li>An image is worth 1000 words </li></ul><ul><li>Many biological systems rely on vision </li></ul><ul><li>The world is 3D and dynamic </li></ul><ul><li>Cameras and computers are cheap </li></ul><ul><li>… </li></ul>
  20. 20. Computer Vision Examples <ul><li>Finding People in images </li></ul><ul><li>Problem 1: Given an image I </li></ul><ul><li>Question: Does I contain an image of a person? </li></ul>
  21. 21. “ Yes” Instances
  22. 22. “ No” Instances
  23. 23. Some Computer Vision Topics
  24. 24. Imaging Geometry
  25. 25. Camera Modeling <ul><li>Pinhole Cameras </li></ul><ul><li>Lenses </li></ul><ul><li>Camera Parameters and Calibration </li></ul>
  26. 26. Image Filtering and Enhancing <ul><li>Linear Filters and Convolution </li></ul><ul><li>Image Smoothing </li></ul><ul><li>Edge Detection </li></ul><ul><li>Pyramids </li></ul>
  27. 27. Image Filtering and Enhancing (cont.)
  28. 28. Region Segmentation
  29. 29. Color
  30. 30. Texture
  31. 31. Image Restoration Original Synthetic
  32. 32. Perceptual Organization
  33. 33. Perceptual Organization
  34. 34. Shape Analysis
  35. 35. Stereo
  36. 36. Motion and Optical Flow
  37. 37. High Level Vision
  38. 38. Image Mosaic
  39. 39. One Very Successful Example <ul><li>Face detection in a digital camera </li></ul><ul><ul><li>The camera detects faces in a scene and then automatically focuses (AF) and optimizes exposure (AE) and, if needed, flash output. </li></ul></ul>
  40. 40. Outline <ul><li>Introductions </li></ul><ul><li>Administrative Matters </li></ul><ul><li>Course Outline </li></ul><ul><li>Applications of Computer Vision </li></ul><ul><li>Computer Vision Focus </li></ul><ul><li>Computer Vision Publications </li></ul>
  41. 41. Applications <ul><li>autonomous cars, planes, missiles, robots, ... </li></ul><ul><li>space exploration </li></ul><ul><li>aid to the blind, ASL recognition </li></ul><ul><li>manufacturing, quality control </li></ul><ul><li>surveillance, security, biometrics </li></ul><ul><li>image retrieval </li></ul><ul><li>medical imaging and analysis </li></ul><ul><li>... </li></ul>
  42. 42. Outline <ul><li>Introductions </li></ul><ul><li>Administrative Matters </li></ul><ul><li>Course Outline </li></ul><ul><li>Applications of Computer Vision </li></ul><ul><li>Computer Vision Focus </li></ul><ul><li>Computer Vision Publications </li></ul>
  43. 43. Computer Vision focuses on: <ul><li>What information should be extracted? </li></ul><ul><li>How can it be extracted? </li></ul><ul><li>How should it be represented? </li></ul><ul><li>How can it be used to achieve the goal? </li></ul>
  44. 44. Related disciplines <ul><li>Image processing </li></ul><ul><li>Pattern recognition </li></ul><ul><li>Photogrammetry </li></ul><ul><li>Computer graphics </li></ul><ul><li>Artificial intelligence </li></ul><ul><li>Machine learning </li></ul><ul><li>Projective geometry </li></ul><ul><li>Control theory </li></ul>
  45. 45. Active Research Topics <ul><li>Object recognition </li></ul><ul><li>Human behavior analysis </li></ul><ul><li>Internet and computer vision </li></ul><ul><li>Biometrics and soft biometrics </li></ul><ul><li>Large scale 3D reconstruction (city level) </li></ul><ul><li>Medical image processing </li></ul><ul><li>Vision for robotics </li></ul><ul><li>… </li></ul>
  46. 46. Outline <ul><li>Introductions </li></ul><ul><li>Administrative Matters </li></ul><ul><li>Course Outline </li></ul><ul><li>Applications of Computer Vision </li></ul><ul><li>Computer Vision Focus </li></ul><ul><li>Computer Vision Publications </li></ul>
  47. 47. Computer Vision Publications <ul><li>Journals </li></ul><ul><ul><li>IEEE Trans. on Pattern Analysis and Machine Intelligence (TPAMI) </li></ul></ul><ul><ul><ul><li>#1 IEEE, Thompson-ISI impact factor: 5.96 </li></ul></ul></ul><ul><ul><ul><li>#1 in both electrical engineering and artificial intelligence </li></ul></ul></ul><ul><ul><ul><li>#3 in all of computer science </li></ul></ul></ul><ul><ul><li>Internal Journal of Computer Vision (IJCV) </li></ul></ul><ul><ul><ul><li>ISI impact factor: 5.358, Rank 2 of 94 in “CS, artificial intelligence </li></ul></ul></ul><ul><ul><li>IEEE Trans. on Image Processing </li></ul></ul><ul><ul><li>… </li></ul></ul>
  48. 48. Importance of CV <ul><li>From these major journal rankings, we can see the importance of Computer Vision research in the whole areas of </li></ul><ul><ul><li>Computer Science </li></ul></ul><ul><ul><li>Electrical Engineering </li></ul></ul>
  49. 49. Computer Vision Publications <ul><li>Conferences </li></ul><ul><ul><li>International Conference on Computer Vision (ICCV) </li></ul></ul><ul><ul><li>Conf. of Computer Vision and Pattern Recognition (CVPR) </li></ul></ul><ul><ul><li>Europe Conference on Computer Vision (ECCV) </li></ul></ul><ul><ul><li>… </li></ul></ul>
  50. 50. <ul><li>Discussions and Questions </li></ul>