Related Experiment Videos
Extraction of motion parallax structure in the visual system. II
1Buys Ballot Laboratory, Utrecht University, The Netherlands.
Biological Cybernetics
|January 1, 1990
Summary
This study presents an algorithm for extracting motion parallax structure from images. Computational results demonstrate its effectiveness using two images, highlighting impacts of discretization and noise.
Area of Science:
- Computer Vision
- Image Processing
- Computational Geometry
Background:
- Motion parallax is crucial for 3D scene understanding.
- Extracting motion structure from image sequences is a fundamental problem.
- Varying image irradiance poses challenges for motion analysis.
Purpose of the Study:
- To present computational results of an algorithm for extracting local affine motion parallax structure.
- To evaluate algorithm performance using only two successive images.
- To analyze the influence of spatio-temporal discretization and internal noise.
Main Methods:
- Developed an algorithm to extract affine motion parallax structure.
- Applied the algorithm to two-image motion sequences.
- Simulated varying image irradiance patterns.
Main Results:
- The algorithm successfully extracts affine motion parallax structure from two images.
- Spatio-temporal discretization significantly affects algorithm performance.
- Internal noise degrades the accuracy of motion structure extraction.
Conclusions:
- The developed algorithm is effective for motion parallax extraction from limited image data.
- Discretization and noise are critical factors to consider for real-world applications.
- The findings are applicable to two-image motion analysis and stereoscopic vision.