ComputerVisionStory [Compatibility Mode]

ComputerVisionStory [Compatibility Mode] - Computer Vision...

Info iconThis preview shows pages 1–13. Sign up to view the full content.

View Full Document Right Arrow Icon
1 Computer Vision Story Mubarak Shah Computer Vision • Computer Vision deals with recovery and use of information about objects present in a scene from images of the scene.
Background image of page 1

Info iconThis preview has intentionally blurred sections. Sign up to view the full version.

View Full DocumentRight Arrow Icon
2 Computer Vision • Computer Vision emerged from: – Image Processing – Pattern Recognition Computer Vision • Computer Vision started as an AI problem.
Background image of page 2
3 AI • Artificial Intelligence is the study of mental faculties through the use of computational faculties through the use of computational models. – Search –NLU – Speech Recognition – Games Games – Computer Vision – Expert Systems Image Understanding • To understand a single image of a scene, locate and identify objects, their structure, and spatial arrangements, and relationships with other objects.
Background image of page 3

Info iconThis preview has intentionally blurred sections. Sign up to view the full version.

View Full DocumentRight Arrow Icon
4 Different Levels • Low Level: Extraction of symbolic information • Intermediate Level • High Level: Interpretation High Level Vision • Image Understanding • Scene Interpretation • Line Drawings
Background image of page 4
5 Interpretation of Line Drawing MIT Copy Demo
Background image of page 5

Info iconThis preview has intentionally blurred sections. Sign up to view the full version.

View Full DocumentRight Arrow Icon
6 What happened? • In order to do line interpretation, need to extract lines from images – Horn-Binford line finder – Solve low level problems before high level problems can be solved. Horn: Physics Based Vision •O p t
Background image of page 6
7 Marr Approach Human vision system Shape from X: Recover 3-D from 2-D Quantitative vs Qualitative Shape from X • Shading •S te reo •Tex
Background image of page 7

Info iconThis preview has intentionally blurred sections. Sign up to view the full version.

View Full DocumentRight Arrow Icon
8 Shape from Texture Shape from Shading
Background image of page 8
9 Shape from Stereo Marr’s Three Levels • Primal Sketch – Marr-Hildreth edge detector •2 . 5 S k e t c h – Marr-Poggio stereo algorithm
Background image of page 9

Info iconThis preview has intentionally blurred sections. Sign up to view the full version.

View Full DocumentRight Arrow Icon
10 After 30 Years of Research • Stereo is almost a solved problem • Structure from motion is very hard • Shape from shading is not interesting/applicable • Range images did not help much • Not much progress in understanding/recognition/interpretation Motion-Based Recognition • A longer sequence leads to recognition of higher level motions like walking or running which level motions, like walking or running,which consist of a complex and coordinated series of events that cannot be understood by looking at only a few frames. – 3-D is not necessary for recognition – Use motion directly for recognition vs Use motion directly for recognition vs • Recognition followed by reconstruction
Background image of page 10
11 Video Understanding •G e s u r e s Activities Activities • Facial expressions • Visual Speech • Applications – Video Surveillance and Monitoring – Perceptual User Interface Perceptual User Interface
Background image of page 11

Info iconThis preview has intentionally blurred sections. Sign up to view the full version.

View Full DocumentRight Arrow Icon
A picture is worth a thousand words. A word is worth a thousand
Background image of page 12
Image of page 13
This is the end of the preview. Sign up to access the rest of the document.

Page1 / 70

ComputerVisionStory [Compatibility Mode] - Computer Vision...

This preview shows document pages 1 - 13. Sign up to view the full document.

View Full Document Right Arrow Icon
Ask a homework question - tutors are online