metadata

library_name: birefnet
tags:
  - background-removal
  - mask-generation
  - Dichotomous Image Segmentation
  - Camouflaged Object Detection
  - Salient Object Detection
  - pytorch_model_hub_mixin
  - model_hub_mixin
repo_url: https://github.com/ZhengPeng7/BiRefNet
pipeline_tag: image-segmentation
license: mit

This BiRefNet was trained with images in 512x512 for faster and more accurate lower resolution inference.

Performance:

All tested in FP16 mode.

Dataset	Method	Resolution	maxFm	wFmeasure	MAE	Smeasure	meanEm	HCE	maxEm	meanFm	adpEm	adpFm	mBA	maxBIoU	meanBIoU
DIS-VD	BiRefNet_512x512-epoch_216	512x512	.879	.840	.040	.888	.931	1526	.941	.864	.938	.857	.732	.747	.726
DIS-VD	BiRefNet-general-epoch_244	512x512	.834	.789	.050	.860	.891	1589	.905	.817	.902	.816	.708	.698	.669
DIS-VD	BiRefNet_HR-general-epoch_130	512x512	.540	.409	.112	.634	.565	1647	.682	.428	.690	.576	.585	.384	.309

Bilateral Reference for High-Resolution Dichotomous Image Segmentation

Peng Zheng^1,4,5,6, Dehong Gao², Deng-Ping Fan^1*, Li Liu³, Jorma Laaksonen⁴, Wanli Ouyang⁵, Nicu Sebe⁶

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

DIS-Sample_1	DIS-Sample_2

This repo is the official implementation of "Bilateral Reference for High-Resolution Dichotomous Image Segmentation" (CAAI AIR 2024).

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 @freepik for their generous support on GPU resources for training this model!

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}
}