Readme
https://hf-mirror.com/ZhengPeng7/BiRefNet-massive
“This BiRefNet for general foreground object extraction is trained on massive data: DIS-TR+DIS-TEs, TR-HRS10K+TE-HRS10K, TR-HRSOD+TE-HRSOD, TR-UHRSD+TE-UHRSD and validated on DIS-VD.”
men1scus
/
birefnet
Bilateral Reference for High-Resolution Dichotomous Image Segmentation (arXiv 2024)
Input
Output
Examples
Run time and cost
This model runs on Nvidia T4 GPU hardware.