Nonlinear component analysis as a kernel eigenvalue problem
Neural Computation
Mixtures of probabilistic principal component analyzers
Neural Computation
Independent component analysis: algorithms and applications
Neural Networks
Controlled animation of video sprites
Proceedings of the 2002 ACM SIGGRAPH/Eurographics symposium on Computer animation
Motion texture: a two-level statistical model for character motion synthesis
Proceedings of the 29th annual conference on Computer graphics and interactive techniques
Interactive motion generation from examples
Proceedings of the 29th annual conference on Computer graphics and interactive techniques
Interactive control of avatars animated with human motion data
Proceedings of the 29th annual conference on Computer graphics and interactive techniques
Getting There: The Ten Top Problems Left
IEEE Computer Graphics and Applications
Laplacian Eigenmaps for dimensionality reduction and data representation
Neural Computation
Information Theory, Inference & Learning Algorithms
Information Theory, Inference & Learning Algorithms
A spatio-temporal extension to Isomap nonlinear dimension reduction
ICML '04 Proceedings of the twenty-first international conference on Machine learning
ACM SIGGRAPH 2005 Papers
Incremental Nonlinear Dimensionality Reduction by Manifold Learning
IEEE Transactions on Pattern Analysis and Machine Intelligence
| Hi-index | 12.05 |
Video texture, a novel type of medium, can produce a new video with a continuously varying stream of images from a recorded video. A classic approach to generate video textures is to apply principal components analysis (PCA) for dimensionality reduction (i.e. extraction of frame signatures) and autoregressive (AR) process for prediction purposes. In this paper we investigate the use of other dimensionality reduction techniques to generate accurate video textures. Based on our experiments, the quality of video textures can be improved further. We also propose a new approach for generating video textures using probabilistic principal components analysis (PPCA) and Gaussian process dynamical model (GPDM) to synthesize video textures which contain frames that never appeared before and with similar motions as original videos. Furthermore, we propose two ways of generating online video textures by applying the incremental Isomap and incremental Spatio-temporal Isomap (IST-Isomap). Both approaches can produce good online video texture results. In particular, IST-Isomap, that we propose, is more suitable for sparse video data (e.g. cartoon).