EECS 598-01: Foundations of Computer Vision
| Instructor: | Jason Corso (jjcorso) |
| Time/Place: | MW 1200-1330 in 1005 DOW |
| Office Hours: | R 0920-1030 in 4227 EECS by appt. |
| Syllabus: | /~jjcorso/t/598F15/files/598F15_syllabus.pdf |
Course Description
Computer Vision seeks to extract useful information from images of various types. This course covers the foundations of computer vision. It emphasizes computer vision as a search for visual invariants and computer vision as mathematical modeling. Foundational representations of images and image content will be discussed. Cross-cutting problems of reduction (e.g., feature extraction, segmentation), estimation (e.g., post estimation, camera calibration), and matching (e.g., image stitching, stereo reconstruction) will be concretely defined and elaborated through many real examples from modern computer vision.
The course is designed for graduate students and will prepare them for future work in computer vision. The students will develop a rich, capable computer vision system over the course of the semester through homework assignments. The students will also produce a term project that reproduces a recent paper in the computer vision literature.
Rough Topic Outline
- Representation and Computer Vision as Optimization
- Visual Invariance (photometric, geometric (shift, rotation, affine, scale), projective, structural, deformation, rate)
- Reduction Problems (feature extraction, segmentation and grouping, etc.)
- Estimation Problems (line and curve fitting, pose estimation, camera calibration, bundle adjustment, etc.)
- Matching Problems (stereo correspondence, image stitching, image registration, etc.)
Other Pieces of Information
The course is designed for graduate students and will prepare them for future work in computer vision. The students will develop a rich, capable computer vision system over the course of the semester through homework assignments. The students will also produce a term project that reproduces a recent paper in the computer vision literature. No prior computer vision course is assumed.The course flyer is here. The course will cater broadly to students across the College of Engineering, LSA (Stats, Math) and School of Information.