Bilateral Reference for High-Resolution Dichotomous Image Segmentation

Peng Zheng 1,4,5,6,โ€‰ Dehong Gao 2,โ€‰ Deng-Ping Fan 1*,โ€‰ Li Liu 3,โ€‰ Jorma Laaksonen 4,โ€‰ Wanli Ouyang 5,โ€‰ Nicu Sebe 6
1 Nankai Universityโ€‚ 2 Northwestern Polytechnical Universityโ€‚ 3 National University of Defense Technologyโ€‚ 4 Aalto Universityโ€‚ 5 Shanghai AI Laboratoryโ€‚ 6 University of Trentoโ€‚
โ€‚ โ€‚ โ€‚ โ€‚ โ€‚ โ€‚ โ€‚ โ€‚ โ€‚

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

Training Sets:

Validation Sets:

  • TE-P3M-500-P

Performance:

Dataset Method Smeasure maxFm meanEm MAE maxEm meanFm wFmeasure adpEm adpFm HCE
TE-P3M-500-P BiRefNet-portrai--epoch_150 .983 .996 .991 .006 .997 .988 .990 .933 .965 .000

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

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

Acknowledgement:

  • Many thanks to @fal for their generous support on GPU resources for training this BiRefNet for portrait matting.

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}
}
Downloads last month
16,437
Safetensors
Model size
221M params
Tensor type
I64
ยท
F32
ยท
Inference Providers NEW
This model isn't deployed by any Inference Provider. ๐Ÿ™‹ Ask for provider support

Model tree for ZhengPeng7/BiRefNet-portrait

Quantizations
1 model

Spaces using ZhengPeng7/BiRefNet-portrait 9