- DensePose task is a part of COCO and Mapillary Joint Recognition Challenge Workshop at ICCV 2019.
See the challenge description and description of the new evaluation metric and the evaluation server.
Dense human pose estimation aims at mapping all human pixels of an RGB image to the 3D surface of the human body.
- We introduce DensePose-COCO, a large-scale ground-truth dataset with image-to-surface correspondences manually annotated on 50K COCO images.
- We propose DensePose-RCNN, a variant of Mask-RCNN, to densely regress part-specific UV coordinates within every human region at multiple frames per second.