Action Recognition Comparison Hub
4 papers - avg viability 6.0
Top Papers
- Human-AI Divergence in Ego-centric Action Recognition under Spatial and Spatiotemporal Manipulations(7.0)
Identify and leverage minimal visual cues for robust action recognition, bridging the gap between human and AI performance in real-world scenarios.
- Novel Semantic Prompting for Zero-Shot Action Recognition(7.0)
SP-CLIP enhances zero-shot action recognition by augmenting vision-language models with structured semantic prompts, offering improved accuracy and efficiency without requiring model retraining.
- Skeleton-to-Image Encoding: Enabling Skeleton Representation Learning via Vision-Pretrained Models(7.0)
Transform skeleton data into images to leverage vision-pretrained models for improved action recognition and representation learning.
- M3GCLR: Multi-View Mini-Max Infinite Skeleton-Data Game Contrastive Learning For Skeleton-Based Action Recognition(3.0)
M3GCLR introduces a game-theoretic approach to enhance skeleton-based action recognition through contrastive learning.