Informative motion extractor for action recognition with kernel feature alignment

This paper proposes a novel algorithm for extracting informative motion features in daily life action recognition based on support vector machine (SVM). The main advantage of the proposed method is not only to extract remarkable motion features, which fit into human intuition, but also to improve the performance of the recognition system. Concretely speaking, the main properties of the proposed method are 1) optimizing kernel parameters so as to minimize its generalization error, 2) extracting remarkable motion features in response to the sensitivity of the kernel function. Experimental result shows that the proposed algorithm improves the accuracy of the recognition system and enables human to identify informative motion features intuitively.

Authors

• Taketoshi Mori
• Masamichi Shimosaka
• Tatsuya Harada
• Tomomasa Sato

Keywords

• Kernel
• Data mining
• Humans
• Feature extraction
• Intelligent robots
• Infrared image sensors
• Legged locomotion
• Information science
• Paper technology
• Intelligent systems
• Action Recognition
• Kernel Feature
• Support Vector
• Support Vector Machine
• Remarkable Feature
• Recognition System
• Motion Features
• Generalization Error
• Kernel Parameters
• Human Intuition
• Walking
• Gradient Descent
• Accuracy Rate
• Sigmoid Function
• Binary Classification
• Feature Space
• Active Targeting
• Input Features
• Step Function
• Detection Scheme
• Human Motion
• Inverse Variance
• Kernel Values
• Input Motion
• Kernel Type
• Recognition Accuracy
• Knowledge Discovery
• Leave-one-out Cross-validation
• Cross-validation Error