A Three-Frame Algorithm for Estimating Two-Component Image Motion
IEEE Transactions on Pattern Analysis and Machine Intelligence
Performance of optical flow techniques
International Journal of Computer Vision
Recursive Filters for Optical Flow
IEEE Transactions on Pattern Analysis and Machine Intelligence
Robustly estimating changes in image appearance
Computer Vision and Image Understanding - Special issue on robusst statistical techniques in image understanding
Computing Optical Flow with Physical Models of Brightness Variation
IEEE Transactions on Pattern Analysis and Machine Intelligence
Spatio-Temporal Image Processing: Theory and Scientific Applications
Spatio-Temporal Image Processing: Theory and Scientific Applications
Measurement of Image Velocity
Optimally Rotation-Equivariant Directional Derivative Kernels
CAIP '97 Proceedings of the 7th International Conference on Computer Analysis of Images and Patterns
Combining the Advantages of Local and Global Optic Flow Methods
Proceedings of the 24th DAGM Symposium on Pattern Recognition
Occlusion Detectable Stereo -- Occlusion Patterns in Camera Matrix
CVPR '96 Proceedings of the 1996 Conference on Computer Vision and Pattern Recognition (CVPR '96)
A Linear Model for Simultaneous Estimation of 3D Motion and Depth
MOTION '02 Proceedings of the Workshop on Motion and Video Computing
ICCV '99 Proceedings of the International Conference on Computer Vision-Volume 2 - Volume 2
Range Flow for Varying Illumination
ECCV '08 Proceedings of the 10th European Conference on Computer Vision: Part I
| Hi-index | 0.00 |
Standard optical flow methods for motion or disparity estimation use a brightness constancy constraint equation (BCCE). This BCCE either handles a moving camera imaging a non-moving scene or a fixed camera imaging a moving scene. In this paper a BCCE is developed that can handle instantaneous motion of the camera on a 2D plane normal to the viewing direction and motion of the imaged scene. From the thus acquired up to 5 dimensional data set 3D object motion, 3D surface element position, and -normals can be estimated simultaneously. Experiments using 1d or 2d camera grids and a weighted total least squares (TLS) estimation scheme demonstrate performance in terms of systematic error and noise stability, and show technical implications.