任务 via/with 模型
如 Skeleton-based Action Recognition via Spatial and Temporal Transformer Networks
Skeleton Based Human Action Recognition with Global Context-Aware Attention LSTM Networks
任务 via/with 想法
如 Efficient Video Action Detection with Token Dropout and Context Refinement
模型 for 任务
如 Two-Stream Convolution Augmented Transformer for Human Activity Recognition
模型 for 任务 with 辅助词或想法
如 Multi-Scale 2D Temporal Adjacency Networks for Moment Localization with Natural Language
模型简称：模型 for 任务
如 TDN: Temporal Difference Networks for Efficient Action Recognition
其它类型
如 Attention is All You Need
