DJW cdece0b32a 第一次提交 7 mēneši atpakaļ
..
coco cdece0b32a 第一次提交 7 mēneši atpakaļ
README.md cdece0b32a 第一次提交 7 mēneši atpakaļ

README.md

Top-down integral-regression-based pose estimation

Top-down methods divide the task into two stages: object detection, followed by single-object pose estimation given object bounding boxes. At the 2nd stage, integral regression based methods use a simple integral operation relates and unifies the heatmap and joint regression differentiably, thus obtain the keypoint coordinates given the features extracted from the bounding box area, following the paradigm introduced in Integral Human Pose Regression.

Results and Models

COCO Dataset

Results on COCO val2017 with detector having human AP of 56.4 on COCO val2017 dataset

Model Input Size AP AR Details and Download
ResNet-50+Debias-IPR 256x256 0.675 0.765 resnet_debias_coco.md
ResNet-50+DSNT 256x256 0.674 0.764 resnet_dsnt_coco.md
ResNet-50+IPR 256x256 0.633 0.730 resnet_ipr_coco.md