Segmenter: Transformer for Semantic Segmentation

Open paper
Benchmark Model Rank Results
semantic-segmentation-on-ade20kSeg-L-Mask/16 (MS)#74Validation mIoU: 53.63
semantic-segmentation-on-ade20kSeg-B-Mask/16(MS, ViT-B)#118Validation mIoU: 50.0
semantic-segmentation-on-ade20kSeg-B/8 (MS, ViT-B)#125Validation mIoU: 49.61
semantic-segmentation-on-ade20k-valSeg-L-Mask/16 (MS, ViT-L)#37mIoU: 53.63
semantic-segmentation-on-ade20k-valSeg-B-Mask/16 (MS, ViT-B)#52mIoU: 50.0
semantic-segmentation-on-ade20k-valSeg-B/8 (MS, ViT-B)#54mIoU: 49.61Pixel Accuracy: 83.37
semantic-segmentation-on-pascal-contextSeg-L-Mask/16#12mIoU: 59.0
thermal-image-segmentation-on-rgb-t-glassSegmenter#15MAE: 0.072

Rank counts only results with a code link.