Abstract:Dynamic texture (DT) exhibits statistical stationarity in the spatial domain and stochastic repetitiveness in the temporal dimension, indicating that different frames of DT possess a high similarity correlation that is critical prior knowledge. However, existing methods cannot effectively learn a synthesis model for high-dimensional DT from a small number of training samples. In this article, we propose a novel DT synthesis method, which makes full use of similarity as prior knowledge to address this issue. Our method is based on the proposed kernel similarity embedding, which can not only mitigate the high dimensionality and small sample issues, but also has the advantage of modeling nonlinear feature relationships. Specifically, we first put forward two hypotheses that are essential for the DT model to generate new frames using similarity correlations. Then, we integrate kernel learning and the extreme learning machine into a unified synthesis model to learn kernel similarity embeddings for representing DTs. Extensive experiments on DT videos collected from the Internet and two benchmark datasets, i.e., Gatech Graphcut Textures and Dyntex, demonstrate that the learned kernel similarity embeddings can provide discriminative representations for DTs. Further, our method can preserve the long-term temporal continuity of the synthesized DT sequences with excellent sustainability and generalization. Meanwhile, it effectively generates realistic DT videos with higher speed and lower computation than the current state-of-the-art methods. The code and more synthesis videos are available at our project page https://shiming-chen.github.io/Similarity-page/Similarit.html .

Kernel Learning for Dynamic Texture Synthesis.

Kernelized Similarity Learning and Embedding for Dynamic Texture Synthesis

Dynamic Texture Modeling and Synthesis Using Multi-Kernel Gaussian Process Dynamic Model

A Novel Method of Dynamic Textures Analysis and Synthesis.

Dynamic Texture Synthesis Via Image Reconstruction.

Locally Linear Embedding Based Dynamic Texture Synthesis.

Equiangular Kernel Dictionary Learning with Applications to Dynamic Texture Analysis

A Survey of Dynamic Texture Synthesis Technic Research

Dynamical Textures Modeling Via Joint Video Dictionary Learning

Modeling Dynamic Textured Background

Dynamic texture synthesis using a spatial temporal descriptor

DTSGAN: Learning Dynamic Textures via Spatiotemporal Generative Adversarial Network

Low Rank Matrix Differential Equations As Deep Kernel Principal Component Analysis

Improving Dynamic Texture Recognition with Constraint Subspace Learning

Dynamic Texture Recognition using PDV Hashing and Dictionary Learning on Multi-scale Volume Local Binary Pattern

Dynamic Texture Recognition via Orthogonal Tensor Dictionary Learning

Manifold Based Dynamic Texture Synthesis from Extremely Few Samples

Dynamic texture recognition with video set based collaborative representation

A Fully Bayesian Infinite Generative Model for Dynamic Texture Segmentation

Dynamic Texture Video Classification Using Extreme Learning Machine

A Sequential Subspace Learning Method and Its Application to Dynamic Texture Analysis