원문정보
초록
영어
Video can be regarded as three dimensional spatio-temporal volume, in which human ac-tion is a three dimensional shape (3D shape) surrounded by the spatio-temporal silhouette surface. The type of human action depends on the shape of the silhouette surface. In this pa-per, we proposed a new feature called Oriented Gradient Histogram of Slide Blocks by build-ing dense overlapping spatio-temporal slide blocks to detect the shape of the 3D silhouette surface of the human action. Sparse coding is adopted to represent videos based on the new feature and Random Forest is utilized to classify the types of human actions. Experiments on KTH and Weizmann human action datasets demonstrate that the new feature can describe the spatio-temporal silhouette surface correctly, accordingly recognize the human action types accurately.
목차
1. Introduction
2. Oriented Gradient Histogram of Slide Blocks
2.1. 3D Gradient Definition
2.2. Spatio-Temporal Silhouette
2.3. Slide Blocks Definition
2.4. New Feature Definition
3. Human Action Recognition Scheme based on New Feature
3.1. New Feature Extraction
3.2. Feature’s Representation with Sparse Coding
3.3. Reducing Dimension by Max Pooling
3.4. Human Action Recognition by Classifiers
4. Experiment and Result
4.1. Experiment Setup
4.2. Evaluation and Performance of our Algorithm on KTH
4.3. Evaluation and Performance of our Algorithm on Weizmann
4.4. Comparison among Three Algorithms
5. Conclusions
Acknowledgments
References