%0 Book Section %1 noKey %A Xu, Li %A Chen, Jianing %A Jia, Jiaya %B Computer Vision – ECCV 2008 %D 2008 %E Forsyth, David %E Torr, Philip %E Zisserman, Andrew %I Springer Berlin Heidelberg %K optical_flow segmentation %P 671-684 %R 10.1007/978-3-540-88682-2_51 %T A Segmentation Based Variational Model for Accurate Optical Flow Estimation %U http://dx.doi.org/10.1007/978-3-540-88682-2_51 %V 5302 %X Segmentation has gained in popularity in stereo matching. However, it is not trivial to incorporate it in optical flow estimation due to the possible non-rigid motion problem. In this paper, we describe a new optical flow scheme containing three phases. First, we partition the input images and integrate the segmentation information into a variational model where each of the segments is constrained by an affine motion. Then the errors brought in by segmentation are measured and stored in a confidence map. The final flow estimation is achieved through a global optimization phase that minimizes an energy function incorporating the confidence map. Extensive experiments show that the proposed method not only produces quantitatively accurate optical flow estimates but also preserves sharp motion boundaries, which makes the optical flow result usable in a number of computer vision applications, such as image/video segmentation and editing. %@ 978-3-540-88681-5

@incollection{noKey, abstract = {Segmentation has gained in popularity in stereo matching. However, it is not trivial to incorporate it in optical flow estimation due to the possible non-rigid motion problem. In this paper, we describe a new optical flow scheme containing three phases. First, we partition the input images and integrate the segmentation information into a variational model where each of the segments is constrained by an affine motion. Then the errors brought in by segmentation are measured and stored in a confidence map. The final flow estimation is achieved through a global optimization phase that minimizes an energy function incorporating the confidence map. Extensive experiments show that the proposed method not only produces quantitatively accurate optical flow estimates but also preserves sharp motion boundaries, which makes the optical flow result usable in a number of computer vision applications, such as image/video segmentation and editing.}, added-at = {2013-07-01T12:54:44.000+0200}, author = {Xu, Li and Chen, Jianing and Jia, Jiaya}, biburl = {https://www.bibsonomy.org/bibtex/28430ab622b5e3824ec25b0a5c03cd6d1/alex_ruff}, booktitle = {Computer Vision – ECCV 2008}, description = {A Segmentation Based Variational Model for Accurate Optical Flow Estimation - Springer}, doi = {10.1007/978-3-540-88682-2_51}, editor = {Forsyth, David and Torr, Philip and Zisserman, Andrew}, interhash = {52d6d3c953cef690a00ff6a9a5a6ef74}, intrahash = {8430ab622b5e3824ec25b0a5c03cd6d1}, isbn = {978-3-540-88681-5}, keywords = {optical_flow segmentation}, pages = {671-684}, publisher = {Springer Berlin Heidelberg}, series = {Lecture Notes in Computer Science}, timestamp = {2013-07-01T12:54:44.000+0200}, title = {A Segmentation Based Variational Model for Accurate Optical Flow Estimation}, url = {http://dx.doi.org/10.1007/978-3-540-88682-2_51}, volume = 5302, year = 2008 }