A2J-Transformer: Anchor-to-Joint Transformer Network for 3D Interacting Hand Pose Estimation from a Single RGB Image.
Changlong JiangYang XiaoCunlin WuMingyang ZhangJinghong ZhengZhiguo CaoJoey Tianyi ZhouPublished in: CVPR (2023)
Keyphrases
- distribution network
- image data
- image classification
- image content
- input image
- image analysis
- single image
- fuzzy logic
- segmentation method
- image features
- multiscale
- image representation
- image regions
- image collections
- fault diagnosis
- spatial information
- edge detection
- image pixels
- color images
- image retrieval
- image segmentation
- region of interest
- network structure
- similarity measure
- segmentation algorithm
- gray scale
- peer to peer
- pixel values
- image registration
- video sequences
- high voltage