ZhengPeng7/BiRefNet_lite-anime

Bilateral Reference for High-Resolution Dichotomous Image Segmentation

Peng Zheng ^1,4,5,6,  Dehong Gao ²,  Deng-Ping Fan ^1*,  Li Liu ³,  Jorma Laaksonen ⁴,  Wanli Ouyang ⁵,  Nicu Sebe ⁶

¹ Nankai University  ² Northwestern Polytechnical University  ³ National University of Defense Technology  ⁴ Aalto University  ⁵ Shanghai AI Laboratory  ⁶ University of Trento 

                 

This repo holds the official weights of BiRefNet for general matting.

Training Sets:

• TR-composited-ani-seg_HR, a composited datasets with fg and bg in the https://huggingface.co/datasets/skytnt/anime-segmentation.

Validation Sets:

• TE-composited-ani-seg_HR

Performance:

Dataset	Method	Smeasure	maxFm	meanEm	MSE	maxEm	meanFm	wFmeasure	adpEm	adpFm	HCE	mBA	maxBIoU	meanBIoU
TE-composited-ani-seg_HR	BiRefNet_lite-anime--epoch_30	.987	.997	.992	.002	.998	.990	.992	.991	.993	.000	.924	.959	.932

Check the main BiRefNet_lite model repo for more info and how to use it:
https://huggingface.co/ZhengPeng7/BiRefNet_lite/blob/main/README.md

Also check the GitHub repo of BiRefNet for all things you may want:
https://github.com/ZhengPeng7/BiRefNet

Citation

@article{zheng2024birefnet,
  title={Bilateral Reference for High-Resolution Dichotomous Image Segmentation},
  author={Zheng, Peng and Gao, Dehong and Fan, Deng-Ping and Liu, Li and Laaksonen, Jorma and Ouyang, Wanli and Sebe, Nicu},
  journal={CAAI Artificial Intelligence Research},
  volume = {3},
  pages = {9150038},
  year={2024}
}