1. 吴心筱 wuxinxiao@bit.edu.cn
计算感知
吴心筱

2. 教材及参考书
1. D. Marr. Vision: A Computational Investigation into the Human Representation and Processing of Visual Information, San Francisco: W. H. Freeman， Rick Szeliski. Computer Vision: Algorithms and application. Springer press，2010.

3. What is Computer Vision

4. Optical illusions

5. Algorithms

6. Industrial applications of computer vision

7. Optical character recognition (OCR): reading handwritten postal codes on letters and automatic number plate recognition (ANPR);

8. Machine inspection: rapid parts inspection for quality assurance using stereo vision with specialized illumination to measure tolerances on aircraft wings or auto body parts;

9. Retail: object recognition for automated checkout lanes

10. Medical imaging: registering pre-operative and intra-operative imagery or performing long-term studies of people’s brain morphology as they age

11. Automotive safety: detecting unexpected obstacles such as pedestrians on the street, under conditions where active vision techniques such as radar or lidar do not work well

12. Surveillance: monitoring for intruders, analyzing highway traffic

13. Motion capture (mocap) Fingerprint recognition and biometrics Match move 3D model building (photogrammetry) ……

14. Consumer-level applications of computer vision

15. Stitching: turning overlapping photos into a single seamlessly stitched panorama

16. Exposure bracketing: merging multiple exposures taken under challenging lighting conditions (strong sunlight and shadows) into a single perfectly exposed image

17. Morphing: turning a picture of one of your friends into another, using a seamless morph transition

18. 3D modeling: converting one or more snapshots into a 3D model of the object or person you are photographing

19. Video match move and stabilization Photo-based walkthroughs Face detection Visual authentication ……

20. A brief of history

21. 1970s
What distinguished computer vision from the already existing field of digital image processing was a desire to recover the three-dimensional structure of the world from images and to use this as a stepping stone towards full scene understanding.

22. Some early (1970s) examples of computer vision algorithms

23. Three levels of an information processing system.
David Marr
Three levels of an information processing system.
Three stages of vision.

24. The three levels

25. The three representation stages

26. The three representation stages

27. 1980s
In the 1980s, a lot of attention was focused on more sophisticated mathematical techniques for performing quantitative image and scene analysis.

28. Examples of computer vision algorithms from the 1980s

29. 1990s
A burst of activity in using projective invariants for recognition evolved into a concerted effort to solve the structure from motion problem. Optical flow methods continued to be improved. Multi-view stereo algorithms that produce complete 3D surfaces were also an active topic of that continues to be active today. Image segmentation, a topic which has been active since the earliest days of computer vision, was also an active topic of research, producing techniques based on minimum energy, and mean shift.

30. 1990s
Tracking algorithms also improved a lot, including contour tracking using active contours, such as snakes, particle filters, and level sets, as well as intensity-based (direct) techniques, often applied to tracking faces and whole bodies. Statistical learning techniques started appearing, first in the application of principal component eigenface analysis to face recognition and linear dynamical systems for curve tracking . Perhaps the most notable development in computer vision during this decade was the increased interaction with computer graphics.

31. Examples of computer vision algorithms from the 1990s

32. 2000s
This past decade has continued to see a deepening interplay between the vision and graphics fields. A second notable trend during this past decade has been the emergence of feature-based techniques (combined with learning) for object . Another significant trend from this past decade has been the development of more efficient algorithms for complex global optimization problem.

33. 2000s
The final trend, which now dominates a lot of the visual recognition research in our community, is the application of sophisticated machine learning techniques to computer vision problems. This trend coincides with the increased availability of immense quantities of partially labeled data on the Internet, which makes it more feasible to learn object categories without the use of careful human supervision.

34. Recent examples of computer vision algorithms

35. Assignment
Please read Chapter of “Vision: A Computational Investigation into the Human Representation and Processing of Visual information(2010)” . Please translate the “The Three Levels” part in P.24-P.27.