Animating pictures with stochastic motion textures
ACM SIGGRAPH 2005 Papers
High Resolution Animated Scenes from Stills
IEEE Transactions on Visualization and Computer Graphics
Animating animal motion from still
ACM SIGGRAPH Asia 2008 papers
Pursuing atomic video words by information projection
ACCV'10 Proceedings of the 10th Asian conference on Computer vision - Volume Part II
Dynamic water motion analysis and rendering
ICIAR'05 Proceedings of the Second international conference on Image Analysis and Recognition
2D motion description and contextual motion analysis: issues and new models
SCVMA'04 Proceedings of the First international conference on Spatial Coherence for Visual Motion Analysis
Intrackability: Characterizing Video Statistics and Pursuing Video Representations
International Journal of Computer Vision
Hi-index | 0.00 |
In this paper, we present a generative model for textured motionphenomena, such as falling snow, wavy river and dancing grass, etc.Firstly, we represent an image as a linear superposition of imagebases selected from a generic and over-complete dictionary. Thedictionary contains Gabor bases for point/particle elements andFourier bases for wave-elements. These bases compete to explain theinput images. The transform from a raw image to a base or a tokenrepresentation leads to large dimension reduction. Secondly, weintroduce a unified motion equation to characterize the motion ofthese bases and the interactions between waves and particles, e.g.a ball floating on water. We use statistical learning algorithm toidentify the structure of moving objects and their trajectoriesautomatically. Then novel sequences can be synthesized easily fromthe motion and image models. Thirdly, we replace the dictionary ofGabor and Fourier bases with symbolic sketches (also bases). Withthe same image and motion model, we can render realistic andstylish cartoon animation. In our view, cartoon and sketch aresymbolic visualization of the inner representation for visualperception. The success of the cartoon animation, in turn, suggeststhat our image and motion models capture the essence of visualperception of textured motion.