Commit
·
707a63f
1
Parent(s):
1b2860f
Add complete metric scores of BiRefNet_HR and BiRefNet for general use.
Browse files
README.md
CHANGED
@@ -15,10 +15,14 @@ license: mit
|
|
15 |
|
16 |
> This BiRefNet was trained with images in `2048x2048` for higher resolution inference.
|
17 |
### Performance:
|
18 |
-
|
19 |
-
|
20 |
-
|
|
21 |
-
|
|
|
|
|
|
|
|
|
|
22 |
|
23 |
<h1 align="center">Bilateral Reference for High-Resolution Dichotomous Image Segmentation</h1>
|
24 |
|
|
|
15 |
|
16 |
> This BiRefNet was trained with images in `2048x2048` for higher resolution inference.
|
17 |
### Performance:
|
18 |
+
> All tested in FP16 mode.
|
19 |
+
|
20 |
+
| Dataset | Method | Resolution | maxFm | wFmeasure | MAE | Smeasure | meanEm | HCE | maxEm | meanFm | adpEm | adpFm | mBA | maxBIoU | meanBIoU |
|
21 |
+
| :------: | :------: | :------: | :------: | :------: | :------: | :------: | :------: | :------: | :------: | :------: | :------: | :------: | :------: | :------: | :------: |
|
22 |
+
| DIS-VD | **BiRefNet_HR**-general-epoch_130 | 2048x2048 | .925 | .894 | .026 | .927 | .952 | 811 | .960 | .909 | .944 | .888 | .828 | .837 | .817 |
|
23 |
+
| DIS-VD | **BiRefNet_HR**-general-epoch_130 | 1024x1024 | .876 | .840 | .041 | .893 | .913 | 1348 | .926 | .860 | .930 | .857 | .765 | .769 | .742 |
|
24 |
+
| DIS-VD | [**BiRefNet**-general-epoch_244](https://huggingface.co/ZhengPeng7/BiRefNet) | 2048x2048 | .888 | .858 | .037 | .898 | .934 | 811 | .941 | .878 | .927 | .862 | .802 | .790 | .776 |
|
25 |
+
| DIS-VD | [**BiRefNet**-general-epoch_244](https://huggingface.co/ZhengPeng7/BiRefNet) | 1024x1024 | .908 | .877 | .034 | .912 | .943 | 1128 | .953 | .894 | .944 | .881 | .796 | .812 | .789 |
|
26 |
|
27 |
<h1 align="center">Bilateral Reference for High-Resolution Dichotomous Image Segmentation</h1>
|
28 |
|