End-to-End Human Pose and Mesh Reconstruction with Transformers.
Kevin LinLijuan WangZicheng LiuPublished in: CoRR (2020)
Keyphrases
- end to end
- human pose
- human body
- pose estimation
- visual hull
- action recognition
- body parts
- three dimensional
- human motion
- congestion control
- human detection
- depth images
- human actions
- high resolution
- pedestrian detection
- random fields
- latent variables
- computer vision
- text localization and recognition
- graph structure
- human computer interaction
- feature points
- probabilistic model
- image sequences
- image segmentation
- multimedia