List your company
ResearchFormerly WizWand

Semantic Segmentation on SUN-RGBD 53 (test)

54.6mIoUGeminiFusion
47.42449.28751.1553.013Dec 6, 2023Jan 23, 2024Mar 12, 2024Apr 30, 2024Jun 17, 2024Aug 5, 2024Sep 23, 2024
Updated 8mo ago

Evaluation results

MethodLinks
GeminiFusionBackbone=Swin-L, Params(M)=369.2, Training dataset=SUN-RGBD2024.0954.6
DAT++-BBackbone=DAT++-B, Params(M)=280.0, Sampling steps=3, Training dataset=SUN-RGBD, Scale value s=0.012024.0953.9
DAT++-SBackbone=DAT++-S, Params(M)=172.8, Sampling steps=3, Training dataset=SUN-RGBD, Scale value s=0.012024.0953.7
GeminiFusionBackbone=MiT-B5, Params(M)=137.2, Training dataset=SUN-RGBD2024.0953.3
GeminiFusionBackbone=MiT-B3, Params(M)=75.8, Training dataset=SUN-RGBD2024.0952.7
DAT++-TBackbone=DAT++-T, Params(M)=73.2, Sampling steps=3, Training dataset=SUN-RGBD, Scale value s=0.012024.0952.6
DFormer-LBackbone=DFormer-L, Params(M)=39.0, Training dataset=SUN-RGBD2024.0952.5
CMX (MiT-B5)Backbone=MiT-B5, Params(M)=181.1, Training dataset=SUN-RGBD2024.0952.4
TokenFusionBackbone=MiT-B3, Params(M)=45.9, Training dataset=SUN-RGBD2024.0952.4
CMX (MiT-B4)Backbone=MiT-B4, Params(M)=139.9, Training dataset=SUN-RGBD2024.0952.1
DAT++-B (Cross-eval)Backbone=DAT++-B, Params(M)=280.0, Sampling steps=3, Training dataset=NYUv2, Scale value s=0.012024.0951.4
DFormer-BBackbone=DFormer-B, Params(M)=29.5, Training dataset=SUN-RGBD2024.0951.2
DAT++-S (Cross-eval)Backbone=DAT++-S, Params(M)=172.8, Sampling steps=3, Training dataset=NYUv2, Scale value s=0.012024.0951.2
DAT++-T (Cross-eval)Backbone=DAT++-T, Params(M)=73.2, Sampling steps=3, Training dataset=NYUv2, Scale value s=0.012024.0950.6
DFormer-SBackbone=DFormer-S, Params(M)=18.7, Training dataset=SUN-RGBD2024.0950
CMXModal=RGB-D2023.1249.7
ShareCMP*Modal=RGB-D, Note=w/o PGA and CPALoss2023.1249.7
TCDModal=RGB-D2023.1249.5
SA-GateModal=RGB-D2023.1249.4
NANetModal=RGB-D2023.1248.8
SGNetModal=RGB-D2023.1248.6
ShapeConvModal=RGB-D2023.1248.6
RDF-152Modal=RGB-D2023.1247.7