EMHIFormer: An Enhanced Multi-Hypothesis Interaction Transformer for 3D human pose estimation in video.
Xuezhi XiangKaixu ZhangYulong QiaoAbdulmotaleb El-SaddikPublished in: J. Vis. Commun. Image Represent. (2023)
Keyphrases
- video sequences
- real time
- video data
- multimedia
- video content
- action detection
- video streams
- spatial and temporal
- real time video
- digital video
- video frames
- space time
- human computer interaction
- user interaction
- public space
- human interaction
- multimedia data
- video processing
- fuzzy logic
- video shots
- artificial intelligence
- power system
- computer vision
- video analysis
- video retrieval
- video surveillance
- object detection
- event detection