Multi-task human analysis in still images: 2D/3D pose, depth map, and multi-part segmentation.
Daniel SánchezMarc OliuMeysam MadadiXavier BaróSergio EscaleraPublished in: FG (2019)
Keyphrases
- depth map
- input image
- multi task
- three dimensional
- image data
- image registration
- multi view
- segmentation method
- d objects
- image classification
- depth estimation
- multi task learning
- depth information
- pose estimation
- level set
- feature points
- stereo matching
- lighting conditions
- high resolution
- object recognition
- image segmentation
- data sets