Realistic human action recognition with multimodal feature selection and fusion

Keyphrases

• Realistic Scenario 100%
• Human Action Recognition 100%
• Human Action 100%
• Multimodal Feature Fusion 100%
• Multimodal Feature Selection 100%
• Performance Improvement 33%
• Support Vector Machine 33%
• Fuzzy Integral 33%
• Recognition Performance 33%
• Video-based 33%
• Dynamic Background 33%
• Audio-Video 33%
• Comprehensive Experiment 33%
• Hollywood 33%
• Visual Features 33%
• Action Recognition 33%
• Audio Cues 33%
• Visual Modality 33%
• Context Influence 33%
• Kernel-based Algorithm 33%
• Audio Signal 33%
• Space-time Interest Points 33%
• Scene Context 33%
• Point Descriptor 33%
• Audio Features 33%
• Audio Context 33%
• Generalized multiple Kernel Learning 33%
• Audio-based 33%
• Effective Feature 33%
• Audio Modality 33%
• Action Image 33%

Computer Science

• Feature Selection 100%
• human action recognition 100%
• Feature Extraction 100%
• Recognition Performance 33%
• Recognition Result 33%
• Interest Point 33%
• Action Recognition 33%
• Visual Feature 33%
• Multiple Kernel Learning 33%
• Visual Modality 33%
• Fuzzy Integral algorithm 33%
• Learning Algorithm 33%
• Performance Improvement 33%
• Experimental Result 33%
• Support Vector Machine 33%