Table 5 Comparison of IoU metrics between the proposed method and state-of-the-art methods in robustness experiments
From: A Mamba based vision transformer for fine grained image segmentation of mural figures
IoU | |||||
|---|---|---|---|---|---|
Methods | Other | Face | Headwear | Body | mIoU |
PSPNet | 86.79 | 73.06 | 37.25 | 76.03 | 68.28 |
OCRNet | 83.97 | 76.54 | 39.89 | 72.20 | 68.15 |
Segmenter | 85.94 | 64.00 | 40.35 | 73.67 | 65.99 |
SegFormer | 85.17 | 74.21 | 42.61 | 71.59 | 68.39 |
DDRNet | 80.57 | 65.49 | 37.29 | 65.28 | 62.16 |
SAN | 78.01 | 78.79 | 46.29 | 65.97 | 67.26 |
Ours | 84.94 | 80.09 | 46.67 | 70.83 | 70.63 |
