OCHuman(Occluded Human) Dataset Api

Dataset proposed in "Pose2Seg: Detection Free Human Instance Segmentation" [ProjectPage] [arXiv] @ CVPR2019.

• News! 2019.06.14 Bug fixed: Val/Test annotation split is now matched to our paper, please update!
• News! 2019.04.08 Codes for our paper is available now!

Samples of OCHuman Dataset

This dataset focus on heavily occluded human with comprehensive annotations including bounding-box, humans pose and instance mask. This dataset contains 13360 elaborately annotated human instances within 5081 images. With average 0.573 MaxIoU of each person, OCHuman is the most complex and challenging dataset related to human. Through this dataset, we want to emphasize occlusion as a challenging problem for researchers to study.

Statistics

All the instances in this dataset are annotated by bounding-box. While not all of them have the keypoint/mask annotation. If you want to compare your results with ours in the paper, please use the subset that contains both keypoint and mask annotations (4731 images, 8110 persons).

Note:

• MaxIoU measures the severity of an object being occluded, which means the max IoU with other same category objects in a single image.
• All instances in OCHuman with kpt/mask annotations are suffered by heavy occlusion. (MaxIou > 0.5)

Download Links

• Images (667MB) & Annotations

In the above link, we also provide the coco style annotations (val and test subset) so that you can run evaluation using cocoEval toolbox.

Update at 2019.06.14: Please download annotation files (*json) again to match the val/test split used in our paper.

Install API

git clone https://github.com/liruilong940607/OCHumanApi
cd OCHumanApi
make install

How to use

See Demo.ipynb