TemPose: a new skeleton-based transformer model designed for fine-grained motion recognition in badminton

Keyphrases

• Motion Recognition 100%
• Transformer Model 100%
• Skeleton-based 100%
• New Skeleton 100%
• Badminton 100%
• Human Motion 50%
• Human Visual 50%
• Human Action 50%
• Player Actions 50%
• Transformer Layer 50%
• Visual Context 50%
• Attention Map 50%
• Interaction Layer 50%
• Temporal Attention 50%
• Parameter Configuration 50%
• Competitive Results 50%
• Additional Inputs 50%
• Novel Skeleton 50%
• Input Stream 50%
• Recognition Benchmark 50%
• Shuttlecock 50%
• Action Recognition 50%
• Multi-person 50%
• Skeleton Model 50%
• Spatial Transformer Network 50%

Computer Science

• Transformer Model 100%
• Human Motions 50%
• Configuration Parameter 50%
• Baseline Model 50%
• Variable Length 50%
• Input-Stream 50%
• Action Recognition 50%