by Lu Qi, Li Jiang, Shu Liu, Xiaoyong Shen, Jiaya Jia.
This repository has released the training and test set of KINS. The annotation format follows COCO style. The mask can be decoded by COCOAPI.
And the reference code of the method in CVPR 2019 paper 'Amodal Instance Segmentation through KINS Dataset' has been released. The codebase is based on pytorch-detectron. You can see some details from our released code. I am sorry that I can not transform it into the maskrcnn-benchmark with clear version.
If our method and dataset are useful for your research, please consider citing:
@inproceedings{qi2019amodal,
title={Amodal Instance Segmentation With KINS Dataset},
author={Qi, Lu and Jiang, Li and Liu, Shu and Shen, Xiaoyong and Jia, Jiaya},
booktitle={Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition},
pages={3014--3023},
year={2019}
}
Please send email to qqlu1992@gmail.com.