Ego-Motion Estimation and 3D Model Refinement in Scenes with Varying Illumination
WACV-MOTION '05 Proceedings of the IEEE Workshop on Motion and Video Computing (WACV/MOTION'05) - Volume 2 - Volume 02
Joint optical flow estimation, segmentation, and 3D interpretation with level sets
Computer Vision and Image Understanding
IEEE Transactions on Pattern Analysis and Machine Intelligence
| Hi-index | 0.00 |
A hierarchical iterative algorithm is proposed for extractingstructure from two-frame optical flow. The algorithmexploits two facts: one is that in many applications,such as face and gesture recognition, the depth variation ofthe visible surface of an object in a scene is small comparedto the distance between the optical center and the object;the other is that the time aliasing problem is alleviated atthe coarse level for any two-frame optical flow estimate sothat the estimate tends to be more accurate. A hierarchicalrepresentation for the relationship between the opticalflow, depth, and the motion parameters is derived, and theresulting non-linear system is iteratively solved through twolinear subsystems. At the coarsest level, the surface of theobject tends to be flat so that the inverse depth tends to bea constant, which is used as the initial depth map. Inversedepth and motion parameters are estimated by the two linearsubsystems at each level and the results are propagatedto finer levels. Error analysis and experiments using bothcomputer-rendered images and real images demonstrate thecorrectness and effectiveness of our algorithm.