Hugging Face's logo

ZhengPeng7
/
BiRefNet_dynamic-matting

Image Segmentation
BiRefNet
Safetensors
Transformers
Transformers.js
background-removal
mask-generation
Dichotomous Image Segmentation
Camouflaged Object Detection
Salient Object Detection
pytorch_model_hub_mixin
model_hub_mixin
custom_code
Model card Files Community
ZhengPeng7 commited on
Commit
c6a7fe9
·
verified ·
1 Parent(s): 8bddc94

Add quantitative comparison of this model for dynamic input shapes and previous ones for input with fixed shapes.

Browse files
Files changed (1) hide show
  1. README.md +5 -1
README.md CHANGED
@@ -20,7 +20,11 @@ license: mit
20
  > This model was trained on arbitrary shapes (256x256 ~ 2304x2304) and shows great robustness on inputs with any shape.
21
 
22
  ### Performance
23
- > How it looks when compared with BiRefNet-matting (fixed 1024x1024 resolution) -- greater than BiRefNet-matting and BiRefNet_HR-matting on the reserved validation sets (DIS-VD and TE-P3M-500-NP).
 
 
 
 
24
 
25
  For performance of different epochs, check the [eval_results-xxx folder for it](https://drive.google.com/drive/u/0/folders/1wSOe0m98YJBRnOefQrC6iefFmeUPtVhn) on my google drive.
26
 
 
20
  > This model was trained on arbitrary shapes (256x256 ~ 2304x2304) and shows great robustness on inputs with any shape.
21
 
22
  ### Performance
23
+ > How it looks when compared with BiRefNet-matting and BiRefNet_HR-matting (fixed resolution, e.g., 1024x1024, 2048x2048).
24
+
25
+ ![comparison_TE-AM-2k](https://cdn-uploads.huggingface.co/production/uploads/63e4977454f51ea342d54814/Mvkf1AWKIfcnpLC6sIgjd.png)
26
+
27
+ ![comparison_TE-P3M-500-NP](https://cdn-uploads.huggingface.co/production/uploads/63e4977454f51ea342d54814/CuACZhbWNN18qIzV5Lzce.png)
28
 
29
  For performance of different epochs, check the [eval_results-xxx folder for it](https://drive.google.com/drive/u/0/folders/1wSOe0m98YJBRnOefQrC6iefFmeUPtVhn) on my google drive.
30