CMSC 426: Image Processing (Computer Vision) David Jacobs.
-
Upload
leslie-lang -
Category
Documents
-
view
223 -
download
0
Transcript of CMSC 426: Image Processing (Computer Vision) David Jacobs.
![Page 1: CMSC 426: Image Processing (Computer Vision) David Jacobs.](https://reader036.fdocuments.in/reader036/viewer/2022081513/56649dc05503460f94ab4ea9/html5/thumbnails/1.jpg)
CMSC 426: Image Processing (Computer Vision)
David Jacobs
![Page 2: CMSC 426: Image Processing (Computer Vision) David Jacobs.](https://reader036.fdocuments.in/reader036/viewer/2022081513/56649dc05503460f94ab4ea9/html5/thumbnails/2.jpg)
Vision
• ``to know what is where, by looking.’’ (Marr).
• Where
• What
![Page 3: CMSC 426: Image Processing (Computer Vision) David Jacobs.](https://reader036.fdocuments.in/reader036/viewer/2022081513/56649dc05503460f94ab4ea9/html5/thumbnails/3.jpg)
Why is Vision Interesting?
• Psychology– ~ 50% of cerebral cortex is for vision.– Vision is how we experience the world.
• Engineering– Want machines to interact with world.– Digital images are everywhere.
![Page 4: CMSC 426: Image Processing (Computer Vision) David Jacobs.](https://reader036.fdocuments.in/reader036/viewer/2022081513/56649dc05503460f94ab4ea9/html5/thumbnails/4.jpg)
Vision is inferential: Light
(http://www-bcs.mit.edu/people/adelson/checkershadow_illusion.html)
![Page 5: CMSC 426: Image Processing (Computer Vision) David Jacobs.](https://reader036.fdocuments.in/reader036/viewer/2022081513/56649dc05503460f94ab4ea9/html5/thumbnails/5.jpg)
Vision is Inferential
![Page 6: CMSC 426: Image Processing (Computer Vision) David Jacobs.](https://reader036.fdocuments.in/reader036/viewer/2022081513/56649dc05503460f94ab4ea9/html5/thumbnails/6.jpg)
Vision is Inferential: Geometry
movie
![Page 7: CMSC 426: Image Processing (Computer Vision) David Jacobs.](https://reader036.fdocuments.in/reader036/viewer/2022081513/56649dc05503460f94ab4ea9/html5/thumbnails/7.jpg)
Vision is Inferential: Prior Knowledge
![Page 8: CMSC 426: Image Processing (Computer Vision) David Jacobs.](https://reader036.fdocuments.in/reader036/viewer/2022081513/56649dc05503460f94ab4ea9/html5/thumbnails/8.jpg)
Vision is Inferential: Prior Knowledge
![Page 9: CMSC 426: Image Processing (Computer Vision) David Jacobs.](https://reader036.fdocuments.in/reader036/viewer/2022081513/56649dc05503460f94ab4ea9/html5/thumbnails/9.jpg)
Computer Vision
• Inference Computation
• Building machines that see
• Modeling biological perception
![Page 10: CMSC 426: Image Processing (Computer Vision) David Jacobs.](https://reader036.fdocuments.in/reader036/viewer/2022081513/56649dc05503460f94ab4ea9/html5/thumbnails/10.jpg)
A Quick Tour of Computer Vision
![Page 11: CMSC 426: Image Processing (Computer Vision) David Jacobs.](https://reader036.fdocuments.in/reader036/viewer/2022081513/56649dc05503460f94ab4ea9/html5/thumbnails/11.jpg)
Boundary Detection: Local cues
![Page 12: CMSC 426: Image Processing (Computer Vision) David Jacobs.](https://reader036.fdocuments.in/reader036/viewer/2022081513/56649dc05503460f94ab4ea9/html5/thumbnails/12.jpg)
Boundary Detection: Local cues
![Page 13: CMSC 426: Image Processing (Computer Vision) David Jacobs.](https://reader036.fdocuments.in/reader036/viewer/2022081513/56649dc05503460f94ab4ea9/html5/thumbnails/13.jpg)
![Page 14: CMSC 426: Image Processing (Computer Vision) David Jacobs.](https://reader036.fdocuments.in/reader036/viewer/2022081513/56649dc05503460f94ab4ea9/html5/thumbnails/14.jpg)
![Page 15: CMSC 426: Image Processing (Computer Vision) David Jacobs.](https://reader036.fdocuments.in/reader036/viewer/2022081513/56649dc05503460f94ab4ea9/html5/thumbnails/15.jpg)
Boundary Detection
http://www.robots.ox.ac.uk/~vdg/dynamics.html
![Page 16: CMSC 426: Image Processing (Computer Vision) David Jacobs.](https://reader036.fdocuments.in/reader036/viewer/2022081513/56649dc05503460f94ab4ea9/html5/thumbnails/16.jpg)
(Sharon, Balun, Brandt, Basri)
![Page 17: CMSC 426: Image Processing (Computer Vision) David Jacobs.](https://reader036.fdocuments.in/reader036/viewer/2022081513/56649dc05503460f94ab4ea9/html5/thumbnails/17.jpg)
![Page 18: CMSC 426: Image Processing (Computer Vision) David Jacobs.](https://reader036.fdocuments.in/reader036/viewer/2022081513/56649dc05503460f94ab4ea9/html5/thumbnails/18.jpg)
Boundary Detection
Finding the Corpus Callosum
(G. Hamarneh, T. McInerney, D. Terzopoulos)
![Page 19: CMSC 426: Image Processing (Computer Vision) David Jacobs.](https://reader036.fdocuments.in/reader036/viewer/2022081513/56649dc05503460f94ab4ea9/html5/thumbnails/19.jpg)
Texture
Photo
Pattern Repeated
![Page 20: CMSC 426: Image Processing (Computer Vision) David Jacobs.](https://reader036.fdocuments.in/reader036/viewer/2022081513/56649dc05503460f94ab4ea9/html5/thumbnails/20.jpg)
Texture
Computer Generated
Photo
![Page 21: CMSC 426: Image Processing (Computer Vision) David Jacobs.](https://reader036.fdocuments.in/reader036/viewer/2022081513/56649dc05503460f94ab4ea9/html5/thumbnails/21.jpg)
Tracking
(Comaniciu and Meer)
![Page 22: CMSC 426: Image Processing (Computer Vision) David Jacobs.](https://reader036.fdocuments.in/reader036/viewer/2022081513/56649dc05503460f94ab4ea9/html5/thumbnails/22.jpg)
Tracking
(www.brickstream.com)
![Page 23: CMSC 426: Image Processing (Computer Vision) David Jacobs.](https://reader036.fdocuments.in/reader036/viewer/2022081513/56649dc05503460f94ab4ea9/html5/thumbnails/23.jpg)
Tracking
![Page 24: CMSC 426: Image Processing (Computer Vision) David Jacobs.](https://reader036.fdocuments.in/reader036/viewer/2022081513/56649dc05503460f94ab4ea9/html5/thumbnails/24.jpg)
Tracking
![Page 25: CMSC 426: Image Processing (Computer Vision) David Jacobs.](https://reader036.fdocuments.in/reader036/viewer/2022081513/56649dc05503460f94ab4ea9/html5/thumbnails/25.jpg)
Tracking
![Page 26: CMSC 426: Image Processing (Computer Vision) David Jacobs.](https://reader036.fdocuments.in/reader036/viewer/2022081513/56649dc05503460f94ab4ea9/html5/thumbnails/26.jpg)
Tracking
![Page 27: CMSC 426: Image Processing (Computer Vision) David Jacobs.](https://reader036.fdocuments.in/reader036/viewer/2022081513/56649dc05503460f94ab4ea9/html5/thumbnails/27.jpg)
Stereo
http://www.ai.mit.edu/courses/6.801/lect/lect01_darrell.pdf
![Page 28: CMSC 426: Image Processing (Computer Vision) David Jacobs.](https://reader036.fdocuments.in/reader036/viewer/2022081513/56649dc05503460f94ab4ea9/html5/thumbnails/28.jpg)
Stereo
http://www.magiceye.com/
![Page 29: CMSC 426: Image Processing (Computer Vision) David Jacobs.](https://reader036.fdocuments.in/reader036/viewer/2022081513/56649dc05503460f94ab4ea9/html5/thumbnails/29.jpg)
Stereo
http://www.magiceye.com/
![Page 30: CMSC 426: Image Processing (Computer Vision) David Jacobs.](https://reader036.fdocuments.in/reader036/viewer/2022081513/56649dc05503460f94ab4ea9/html5/thumbnails/30.jpg)
Motion
http://www.ai.mit.edu/courses/6.801/lect/lect01_darrell.pdf
![Page 31: CMSC 426: Image Processing (Computer Vision) David Jacobs.](https://reader036.fdocuments.in/reader036/viewer/2022081513/56649dc05503460f94ab4ea9/html5/thumbnails/31.jpg)
Motion - Application
(www.realviz.com)
![Page 32: CMSC 426: Image Processing (Computer Vision) David Jacobs.](https://reader036.fdocuments.in/reader036/viewer/2022081513/56649dc05503460f94ab4ea9/html5/thumbnails/32.jpg)
Pose Determination
Visually guided surgery
![Page 33: CMSC 426: Image Processing (Computer Vision) David Jacobs.](https://reader036.fdocuments.in/reader036/viewer/2022081513/56649dc05503460f94ab4ea9/html5/thumbnails/33.jpg)
Recognition - Shading
Lighting affects appearance
![Page 34: CMSC 426: Image Processing (Computer Vision) David Jacobs.](https://reader036.fdocuments.in/reader036/viewer/2022081513/56649dc05503460f94ab4ea9/html5/thumbnails/34.jpg)
![Page 35: CMSC 426: Image Processing (Computer Vision) David Jacobs.](https://reader036.fdocuments.in/reader036/viewer/2022081513/56649dc05503460f94ab4ea9/html5/thumbnails/35.jpg)
![Page 36: CMSC 426: Image Processing (Computer Vision) David Jacobs.](https://reader036.fdocuments.in/reader036/viewer/2022081513/56649dc05503460f94ab4ea9/html5/thumbnails/36.jpg)
Classification
(Funkhauser, Min, Kazhdan, Chen, Halderman, Dobkin, Jacobs)
![Page 37: CMSC 426: Image Processing (Computer Vision) David Jacobs.](https://reader036.fdocuments.in/reader036/viewer/2022081513/56649dc05503460f94ab4ea9/html5/thumbnails/37.jpg)
Vision depends on:
• Geometry
• Physics
• The nature of objects in the world
(This is the hardest part).
![Page 38: CMSC 426: Image Processing (Computer Vision) David Jacobs.](https://reader036.fdocuments.in/reader036/viewer/2022081513/56649dc05503460f94ab4ea9/html5/thumbnails/38.jpg)
Approaches to Vision
![Page 39: CMSC 426: Image Processing (Computer Vision) David Jacobs.](https://reader036.fdocuments.in/reader036/viewer/2022081513/56649dc05503460f94ab4ea9/html5/thumbnails/39.jpg)
Modeling + Algorithms
• Build a simple model of the world
(eg., flat, uniform intensity).
• Find provably good algorithms.
• Experiment on real world.
• Update model.
Problem: Too often models are simplistic or intractable.
![Page 40: CMSC 426: Image Processing (Computer Vision) David Jacobs.](https://reader036.fdocuments.in/reader036/viewer/2022081513/56649dc05503460f94ab4ea9/html5/thumbnails/40.jpg)
Bayesian inference
• Bayes law: P(A|B) = P(B|A)*P(A)/P(B).• P(world|image) = P(image|world)*P(world)/P(image)• P(image|world) is computer graphics
– Geometry of projection.– Physics of light and reflection.
• P(world) means modeling objects in world. Leads to statistical/learning approaches.Problem: Too often probabilities can’t be known and
are invented.
![Page 41: CMSC 426: Image Processing (Computer Vision) David Jacobs.](https://reader036.fdocuments.in/reader036/viewer/2022081513/56649dc05503460f94ab4ea9/html5/thumbnails/41.jpg)
Engineering
• Focus on definite tasks with clear requirements.
• Try ideas based on theory and get experience about what works.
• Try to build reusable modules.
Problem: Solutions that work under specific conditions may not generalize.
![Page 42: CMSC 426: Image Processing (Computer Vision) David Jacobs.](https://reader036.fdocuments.in/reader036/viewer/2022081513/56649dc05503460f94ab4ea9/html5/thumbnails/42.jpg)
The State of Computer Vision
• Science– Study of intelligence seems to be hard.– Some interesting fundamental theory about
specific problems.– Limited insight into how these interact.
![Page 43: CMSC 426: Image Processing (Computer Vision) David Jacobs.](https://reader036.fdocuments.in/reader036/viewer/2022081513/56649dc05503460f94ab4ea9/html5/thumbnails/43.jpg)
The State of Computer Vision
• Technology– Interesting applications: inspection,
graphics, security, internet….– Some successful companies. Largest
~100-200 million in revenues. Many in-house applications.
– Future: growth in digital images exciting.
![Page 44: CMSC 426: Image Processing (Computer Vision) David Jacobs.](https://reader036.fdocuments.in/reader036/viewer/2022081513/56649dc05503460f94ab4ea9/html5/thumbnails/44.jpg)
Related Fields
• Graphics. “Vision is inverse graphics”.• Visual perception.• Neuroscience.• AI• Learning• Math: eg., geometry, stochastic processes.• Optimization.
![Page 45: CMSC 426: Image Processing (Computer Vision) David Jacobs.](https://reader036.fdocuments.in/reader036/viewer/2022081513/56649dc05503460f94ab4ea9/html5/thumbnails/45.jpg)
Contact Info
Prof: David Jacobs
Office: Room 4421, A.V. Williams Building (Next to CSIC).
Phone: (301) 405-0679 Email: [email protected] Homepage: http://www.cs.umd.edu/~djacobs
TA: Hyoungjune Yi
Email: [email protected]
![Page 46: CMSC 426: Image Processing (Computer Vision) David Jacobs.](https://reader036.fdocuments.in/reader036/viewer/2022081513/56649dc05503460f94ab4ea9/html5/thumbnails/46.jpg)
Tools Needed for Course
• Math– Calculus– Linear Algebra (can be picked up).
• Computer Science– Algorithms– Programming, we’ll use Matlab.
• Signal Processing (we’ll teach a little).
![Page 47: CMSC 426: Image Processing (Computer Vision) David Jacobs.](https://reader036.fdocuments.in/reader036/viewer/2022081513/56649dc05503460f94ab4ea9/html5/thumbnails/47.jpg)
Rough Syllabus
![Page 48: CMSC 426: Image Processing (Computer Vision) David Jacobs.](https://reader036.fdocuments.in/reader036/viewer/2022081513/56649dc05503460f94ab4ea9/html5/thumbnails/48.jpg)
Course Organization
• Reading assignments in Forsyth & Ponce, plus some extras.
• ~6-8 Problem sets - Programming and paper and pencil• Two quizzes, Final Exam.• Grading: Problem sets 30%, quizzes: first quiz 10%;
second quiz 20%; final 40%.• Web page:
www.cs.umd.edu/~djacobs/CMSC426/CMSC426.htm