Dancing Avatar: Pose and Text-Guided Human Motion Videos Synthesis with Image Diffusion Model.
Bosheng QinWentao YeQifan YuSiliang TangYueting ZhuangPublished in: CoRR (2023)
Keyphrases
- human motion
- diffusion model
- video sequences
- image data
- visual data
- input image
- human actions
- image analysis
- human body
- motion synthesis
- image segmentation
- image retrieval
- image sequences
- key frames
- high resolution
- multiscale
- feature points
- motion capture
- diffusion process
- pose estimation
- edge detection
- body parts
- body pose
- medical images
- d objects
- spatio temporal
- human pose
- moving objects
- motion capture data