MMSFormer: Multimodal Transformer for Material and Semantic Segmentation

Open paper
Benchmark Model Rank Results
semantic-segmentation-on-fmb-datasetMMSFormer (RGB-Infrared)#4mIoU: 61.70
semantic-segmentation-on-fmb-datasetMMSFormer (RGB)#5mIoU: 57.20
semantic-segmentation-on-mcubesMMSFormer (RGB-A-D-N)#4mIoU: 53.11%
semantic-segmentation-on-mcubesMMSFormer (RGB-A-D)#9mIoU: 52.05%
semantic-segmentation-on-mcubesMMSFormer (RGB-A)#11mIoU: 51.30%
semantic-segmentation-on-mcubesMMSFormer (RGB)#15mIoU: 50.44%
semantic-segmentation-on-mcubes-pMMSFormer (RGB-A-D)#1mIoU: 52.03
semantic-segmentation-on-mcubes-pMMSFormer (RGB-A)#2mIoU: 51.30
semantic-segmentation-on-mcubes-pMMSFormer (RGB)#5mIoU: 50.44
thermal-image-segmentation-on-pst900MMSFormer#5mIoU: 87.45

Rank counts only results with a code link.