Texture Synthesis Method for Generic Video Sequences

@inproceedings{citeulike:2908435, abstract = {An effective texture synthesis method is presented that is inspired by the work of Kwatra et al. [1]. Their algorithm is non-parametric and patch-based. Blending between overlapping patches is optimized using graph cut techniques. We generalize the initial approach [ 1 ] to achieve a new synthesis algorithm that yields improved results for a much larger class of natural video sequences. For that, two major extensions have been provided: (1) the ability to handle constrained texture synthesis applications and (2) robustness against global camera motion. Constrained synthesis thereby refers to integrating synthetic textures into natural video sequences, as opposed to unconstrained texture synthesis, where (infinite) spatio-temporal extensions of single textures are generated. Camera motion compensation enables applicability of the synthesis algorithm to video sequences with a moving camera. The results presented in this paper show that the proposed improvements yield significant subjective gains compared to the initial algorithm.}, author = {Ndjiki-Nya, P. and Stuber, C. and Wiegand, T. }, booktitle = {Image Processing, 2007. ICIP 2007. IEEE International Conference on}, citeulike-article-id = {2908435}, comment = {This paper is proposing an improvement of the method described by Kwatra et al. for video texture synthesis. The improvement is two-fold: first, the "empty" regions in videos are not randomly filled with texture, but there's a "constraint texture" region introduced so that the transition between the smooth and the synthesized textures is smooth. The second benefit is temporal alignment of the Kwatra algorithm - accounting for the camera motion (which is assumed to be 0 in the initial algorithm).}, doi = {10.1109/ICIP.2007.4379330}, journal = {Image Processing, 2007. ICIP 2007. IEEE International Conference on}, keywords = {synthesis, texture, video}, pages = {III - 397--III - 400}, posted-at = {2008-06-19 19:51:10}, priority = {0}, title = {Texture Synthesis Method for Generic Video Sequences}, url = {http://dx.doi.org/10.1109/ICIP.2007.4379330}, volume = {3}, year = {2007} }

TY - CONF ID - citeulike:2908435 L3 - citeulike-article-id:2908435 TI - Texture Synthesis Method for Generic Video Sequences T2 - Image Processing, 2007. ICIP 2007. IEEE International Conference on JF - Image Processing, 2007. ICIP 2007. IEEE International Conference on VL - 3 SP - III - 397 EP - III - 400 N2 - An effective texture synthesis method is presented that is inspired by the work of Kwatra et al. [1]. Their algorithm is non-parametric and patch-based. Blending between overlapping patches is optimized using graph cut techniques. We generalize the initial approach [ 1 ] to achieve a new synthesis algorithm that yields improved results for a much larger class of natural video sequences. For that, two major extensions have been provided: (1) the ability to handle constrained texture synthesis applications and (2) robustness against global camera motion. Constrained synthesis thereby refers to integrating synthetic textures into natural video sequences, as opposed to unconstrained texture synthesis, where (infinite) spatio-temporal extensions of single textures are generated. Camera motion compensation enables applicability of the synthesis algorithm to video sequences with a moving camera. The results presented in this paper show that the proposed improvements yield significant subjective gains compared to the initial algorithm. KW - synthesis KW - texture KW - video N1 - This paper is proposing an improvement of the method described by Kwatra et al. for video texture synthesis. The improvement is two-fold: first, the "empty" regions in videos are not randomly filled with texture, but there's a "constraint texture" region introduced so that the transition between the smooth and the synthesized textures is smooth. The second benefit is temporal alignment of the Kwatra algorithm - accounting for the camera motion (which is assumed to be 0 in the initial algorithm). AU - Ndjiki-Nya, P AU - Stuber, C AU - Wiegand, T PY - 2007/// UR - http://dx.doi.org/10.1109/ICIP.2007.4379330 ER -