Identity From Here, Pose From There: Self-Supervised Disentanglement and Generation of Objects Using Unlabeled Videos.
Fanyi XiaoHaotian LiuYong Jae LeePublished in: ICCV (2019)
Keyphrases
- d objects
- dynamic scenes
- video images
- pose estimation
- object segmentation
- video sequences
- object appearance
- object representation
- multiple objects
- object classes
- partial occlusion
- unsupervised learning
- moving objects
- training data
- data objects
- video frames
- labeled data
- object model
- multi view
- video analysis
- moving camera
- articulated objects
- video segments
- supervised learning
- planar objects
- object pose