T2M-GPT: Generating Human Motion from Textual Descriptions with Discrete Representations.
Jianrong ZhangYangsong ZhangXiaodong CunShaoli HuangYong ZhangHongwei ZhaoHongtao LuXi ShenPublished in: CoRR (2023)
Keyphrases
- human motion
- textual descriptions
- motion capture
- human body
- video sequences
- image sequences
- motion recognition
- spatio temporal
- motion capture data
- metadata
- semantic concepts
- higher level
- text classification
- semantic information
- visual data
- semantic representation
- image understanding
- co occurrence
- keyword queries
- object detection
- keywords
- database