000100992 001__ 100992
000100992 005__ 20190316233937.0
000100992 0247_ $$2doi$$a10.1109/TIP.2007.910956
000100992 022__ $$a1057-7149
000100992 02470 $$2ISI$$a000251744700005
000100992 02470 $$2DAR$$a11937
000100992 037__ $$aARTICLE
000100992 245__ $$aHigher Order SVD Analysis for Dynamic Texture Synthesis
000100992 269__ $$a2008
000100992 260__ $$bInstitute of Electrical and Electronics Engineers$$c2008
000100992 336__ $$aJournal Articles
000100992 520__ $$aVideos representing flames, water, smoke, etc. are often defined as dynamic textures: "textures" because they are characterized by redundant repetition of a pattern and "dynamic" because this repetition is also in time and not only in space. Dynamic textures have been modeled as linear dynamic systems by unfolding the video frames into column vectors and describing their trajectory as time evolves. After the projection of the vectors onto a lower dimensional space by a Singular Value Decomposition (SVD), the trajectory is modeled using system identification techniques. Synthesis is obtained by driving the system with random noise. In this paper, we show that the standard SVD can be replaced by a Higher Order SVD (HOSVD), originally known as Tucker decomposition. HOSVD decomposes the dynamic texture as a multidimensional signal (tensor) without unfolding the video frames on column vectors. This is a more natural and flexible decomposition, since it permits to perform dimension reduction in spatial, temporal, and chromatic domain, while standard SVD allows for temporal reduction only. We show that for a comparable synthesis quality, the HOSVD approach requires on average five times less parameters than the standard SVD approach. The analysis part is more expensive, but the synthesis has the same cost as existing algorithms. Our technique is thus well suited to dynamic texture synthesis on devices limited by memory and computational power, such as PDAs or mobile phones.
000100992 6531_ $$adynamic texture
000100992 6531_ $$asynthesis
000100992 6531_ $$alinear systems
000100992 6531_ $$ahigher order SVD
000100992 6531_ $$aIVRG
000100992 700__ $$aCostantini, Roberto
000100992 700__ $$0244018$$g115222$$aSbaiz, Luciano
000100992 700__ $$aSüsstrunk, Sabine$$g125681$$0241946
000100992 773__ $$j17$$tIEEE Transactions on Image Processing$$k1$$q42-52
000100992 8564_ $$uhttp://lcavwww.epfl.ch/reproducible_research/CostantiniIP07/index.html$$zURL
000100992 8564_ $$uhttps://infoscience.epfl.ch/record/100992/files/CostantiniSS2008.pdf$$zn/a$$s3207588$$yn/a
000100992 8564_ $$uhttps://infoscience.epfl.ch/record/100992/files/Costantini_RR_IP07_code_rr.zip$$s47878742
000100992 909C0 $$xU10434$$0252056$$pLCAV
000100992 909C0 $$pIVRL$$xU10429$$0252320
000100992 909CO $$qGLOBAL_SET$$pIC$$particle$$ooai:infoscience.tind.io:100992
000100992 917Z8 $$x114218
000100992 917Z8 $$x114218
000100992 917Z8 $$x190838
000100992 937__ $$aLCAV-ARTICLE-2007-006
000100992 973__ $$rREVIEWED$$sPUBLISHED$$aEPFL
000100992 980__ $$aARTICLE