Semantic Segmentation on Trans10K v2 (test)
81.37mIoUOurs-B5
Evaluation Results
| Method | Links | ||||||||||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Ours-B5GFLOPs=154.37, MParams=106.192025.12 | 81.37 | 96.93 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| Ours-B4GFLOPs=79.34, MParams=67.112025.12 | 80.99 | 96.59 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| Ours-B3GFLOPs=68.35, MParams=51.212025.12 | 80.04 | 96.28 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| Ours-B2GFLOPs=37.03, MParams=27.592025.12 | 79.29 | 95.92 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| Ours-B2GFLOPs=37.032025.12 | 79.29 | 95.92 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| To-Former-B2GFLOPs=117.742025.12 | 77.43 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| Ours-LGFLOPs=50.54, MParams=60.862025.12 | 77.35 | 95.28 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| Ours-B1GFLOPs=21.29, MParams=14.872025.12 | 77.05 | 95.37 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| Ours-MGFLOPs=34.51, MParams=43.702025.12 | 76.06 | 95.08 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| Ours-SGFLOPs=20.00, MParams=23.982025.12 | 75.32 | 94.83 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| Trans4Trans-MGFLOPs=34.382025.12 | 75.14 | 95.01 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| Trans4Trans-MGFLOPs=34.382025.12 | 75.14 | 95.01 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| Trans4Trans-MEncoder=PVT-M [67], Decoder=TPM / Single-head, GFLOPs=34.38, MParams=43.652021.08 | 75.14 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| Trans4Trans-MEncoder=PVT-M [63], Decoder=TPM / Single-head, GFLOPs=34.38, MParams=43.65, Input Size=512x5122021.07 | 75.14 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| Trans4Trans-MGFLOPs=34.382021.07 | 75.14 | 95.01 | 96.08 | — | 5,581 | — | 6,925 | 65.16 | 63.96 | 83.84 | 88.21 | 80.29 | 76.33 | 8,309 | 68.09 | 71.46 | |
| Trans4Trans-MEncoder=PVT-M [67], Decoder=TPM / Dual-head, GFLOPs=35.17, MParams=44.042021.08 | 74.98 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| Trans4Trans-MEncoder=PVT-M [63], Decoder=TPM / Dual-head, GFLOPs=35.17, MParams=44.04, Input Size=512x5122021.07 | 74.98 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| Trans4Trans-SEncoder=PVT-S [67], Decoder=TPM / Dual-head, GFLOPs=20.69, MParams=24.342021.08 | 74.57 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| Trans4Trans-SEncoder=PVT-S [63], Decoder=TPM / Dual-head, GFLOPs=20.69, MParams=24.34, Input Size=512x5122021.07 | 74.57 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| Trans4Trans-SGFLOPs=19.922025.12 | 74.15 | 94.57 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| Trans4Trans-SEncoder=PVT-S [67], Decoder=TPM / Single-head, GFLOPs=19.92, MParams=23.952021.08 | 74.15 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| Trans4Trans-SEncoder=PVT-S [63], Decoder=TPM / Single-head, GFLOPs=19.92, MParams=23.95, Input Size=512x5122021.07 | 74.15 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| Trans4Trans-SGFLOPs=19.922021.07 | 74.15 | 94.57 | 95.6 | — | 5,705 | — | 7,021 | 63.95 | 61.25 | 81.67 | 87.34 | 78.52 | 77.13 | 8,100 | 64.88 | 71.18 | |
| Trans2SegFLOPs=49.032021.01 | 72.15 | 94.14 | 95.35 | — | 53.43 | 6,782 | 6,420 | 59.64 | 60.56 | 88.52 | 86.67 | 75.99 | 73.98 | 8,243 | 57.17 | — | |
| Trans2SegGFLOPs=49.03, MParams=56.202025.12 | 72.15 | 94.14 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| Trans2SegGFLOPs=49.032025.12 | 72.15 | 94.14 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| Trans2SegGFLOPs=49.032021.07 | 72.15 | 94.14 | 95.35 | — | 5,343 | — | 6,420 | 59.64 | 60.56 | 88.52 | 86.67 | 75.99 | 73.98 | 8,243 | 57.17 | 67.82 | |
| Trans2Seg-MEncoder=PVT-M [67], Decoder=Transformer [24], GFLOPs=49.00, MParams=56.202021.08 | 72.1 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| PVT-MEncoder=PVT-M [63], Decoder=Transformer [70], GFLOPs=49, MParams=56.2, Input Size=512x5122021.07 | 72.1 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| Trans4Trans-TEncoder=PVT-T [67], Decoder=TPM / Dual-head, GFLOPs=11.22, MParams=13.102021.08 | 69.84 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| Trans4Trans-TEncoder=PVT-T [63], Decoder=TPM / Dual-head, GFLOPs=11.22, MParams=13.1, Input Size=512x5122021.07 | 69.84 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| Ours-TGFLOPs=10.50, MParams=12.722025.12 | 69.53 | 93.52 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| Trans2Seg-SEncoder=R50 [24], Decoder=Transformer [70], GFLOPs=40.98, MParams=30.53, Input Size=512x5122021.07 | 69.2 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| TranslabFLOPs=61.312021.01 | 69 | 92.67 | 93.9 | — | 54.36 | 6,448 | 6,514 | 54.58 | 57.72 | 79.85 | 81.61 | 72.82 | 69.63 | 7,750 | 56.43 | — | |
| TransLabGFLOPs=61.31, MParams=42.192025.12 | 69 | 92.67 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| TransLabGFLOPs=61.312021.07 | 69 | 92.67 | 93.9 | — | 5,436 | — | 6,514 | 54.58 | 57.72 | 79.85 | 81.61 | 72.82 | 69.63 | 7,750 | 56.43 | 64.48 | |
| DeepLabv3+FLOPs=37.982021.01 | 68.87 | 92.75 | 93.82 | — | 51.29 | 6,465 | 6,571 | 55.26 | 57.19 | 77.06 | 81.89 | 72.64 | 70.81 | 7,744 | 58.63 | — | |
| DeepLabv3+GFLOPs=37.98, MParams=28.742025.12 | 68.87 | 92.75 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| DeepLabv3+GFLOPs=37.982025.12 | 68.87 | 92.75 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| DeepLabv3+GFLOPs=37.982021.07 | 68.87 | 92.75 | 93.82 | — | 5,129 | — | 6,571 | 55.26 | 57.19 | 77.06 | 81.89 | 72.64 | 70.81 | 7,744 | 58.63 | 64.65 | |
| Trans2Seg-TEncoder=R34 [24], Decoder=Transformer [70], GFLOPs=30.26, MParams=27.98, Input Size=512x5122021.07 | 68.84 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| DANetFLOPs=198.02021.01 | 68.81 | 92.7 | 93.69 | — | 47.69 | 6,605 | 7,018 | 53.01 | 56.15 | 77.73 | 82.89 | 72.24 | 72.18 | 7,787 | 56.06 | — | |
| DANetGFLOPs=198.002025.12 | 68.81 | 92.7 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| DANetGFLOPs=1982021.07 | 68.81 | 92.7 | 93.69 | — | 4,769 | — | 7,018 | 53.01 | 56.15 | 77.73 | 82.89 | 72.24 | 72.18 | 7,787 | 56.06 | 66.05 | |
| Trans4Trans-TGFLOPs=10.452025.12 | 68.63 | 93.23 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| Trans4Trans-TEncoder=PVT-T [67], Decoder=TPM / Single-head, GFLOPs=10.45, MParams=12.712021.08 | 68.63 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| Trans4Trans-TEncoder=PVT-T [63], Decoder=TPM / Single-head, GFLOPs=10.45, MParams=12.71, Input Size=512x5122021.07 | 68.63 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| Trans4Trans-TGFLOPs=10.452021.07 | 68.63 | 93.23 | 94.44 | — | 4,839 | — | 6,186 | 61.14 | 54.83 | 73.6 | 83.03 | 75.2 | 74.69 | 7,526 | 59.19 | 61.89 | |
| Trans2Seg-SEncoder=PVT-S [67], Decoder=Transformer [24], GFLOPs=19.58, MParams=24.362021.08 | 68.47 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| PVT-SEncoder=PVT-S [63], Decoder=Transformer [70], GFLOPs=19.58, MParams=24.36, Input Size=512x5122021.07 | 68.47 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| PSPNetFLOPs=187.032021.01 | 68.23 | 92.47 | 93.62 | — | 50.33 | 6,424 | 7,019 | 51.51 | 55.27 | 79.27 | 81.93 | 71.95 | 68.91 | 7,713 | 54.43 | — | |
| PSPNetGFLOPs=187.03, MParams=50.992025.12 | 68.23 | 92.47 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| PSPNetGFLOPs=187.032021.07 | 68.23 | 92.47 | 93.62 | — | 5,033 | — | 7,019 | 51.51 | 55.27 | 79.27 | 81.93 | 71.95 | 68.91 | 7,713 | 54.43 | 64.24 | |
| OCNetFLOPs=43.312021.01 | 66.31 | 92.03 | 93.12 | — | 41.47 | 6,354 | 6,005 | 54.1 | 51.01 | 79.57 | 81.95 | 69.4 | 68.44 | 7,841 | 54.65 | — | |
| OCNetGFLOPs=43.312025.12 | 66.31 | 92.03 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| OCNetGFLOPs=43.312021.07 | 66.31 | 92.03 | 93.12 | — | 4,147 | — | 6,005 | 54.1 | 51.01 | 79.57 | 81.95 | 69.4 | 68.44 | 7,841 | 54.65 | 63.54 | |
| Trans2Seg-TEncoder=PVT-T [67], Decoder=Transformer [24], GFLOPs=10.16, MParams=13.112021.08 | 64.6 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| PVT-TEncoder=PVT-T [63], Decoder=Transformer [70], GFLOPs=10.16, MParams=13.11, Input Size=512x5122021.07 | 64.6 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| Trans2Seg-TEncoder=R18 [24], Decoder=Transformer [70], GFLOPs=16.96, MParams=17.87, Input Size=512x5122021.07 | 64.2 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| DenseASPPFLOPs=36.202021.01 | 63.01 | 90.86 | 91.39 | — | 42.41 | 6,093 | 6,475 | 48.97 | 51.4 | 65.72 | 75.64 | 67.93 | 67.03 | 7,026 | 49.64 | — | |
| DenseASPPGFLOPs=36.20, MParams=29.092025.12 | 63.01 | 90.86 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| DenseASPPGFLOPs=36.22025.12 | 63.01 | 90.86 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| DenseASPPGFLOPs=36.22021.07 | 63.01 | 90.86 | 91.39 | — | 4,241 | — | 6,475 | 48.97 | 51.4 | 65.72 | 75.64 | 67.93 | 67.03 | 7,026 | 49.64 | 60.93 | |
| FCNFLOPs=42.232021.01 | 62.75 | 91.65 | 93.62 | — | 38.84 | 5,605 | 5,876 | 46.91 | 50.74 | 82.56 | 78.71 | 68.78 | 57.87 | 7,366 | 46.54 | — | |
| FCNGFLOPs=42.23, MParams=34.992025.12 | 62.75 | 91.65 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| FCNGFLOPs=42.232021.07 | 62.75 | 91.65 | 93.62 | — | 3,884 | — | 5,876 | 46.91 | 50.74 | 82.56 | 78.71 | 68.78 | 57.87 | 7,366 | 46.54 | 56.05 | |
| DUNetFLOPs=123.692021.01 | 59.01 | 90.67 | 93.07 | — | 34.2 | 5,095 | 5,496 | 43.19 | 45.05 | 79.8 | 76.07 | 65.29 | 54.33 | 6,857 | 42.64 | — | |
| DUNetGFLOPs=123.692025.12 | 59.01 | 90.67 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| DUNetGFLOPs=123.692021.07 | 59.01 | 90.67 | 93.07 | — | 3,420 | — | 5,496 | 43.19 | 45.05 | 79.8 | 76.07 | 65.29 | 54.33 | 6,857 | 42.64 | 50.95 | |
| BiSeNetFLOPs=19.912021.01 | 58.4 | 89.13 | 90.12 | — | 39.54 | 5,371 | 5,090 | 46.95 | 44.68 | 64.32 | 72.86 | 63.57 | 61.38 | 6,788 | 44.85 | — | |
| BiSeNetGFLOPs=19.91, MParams=13.32025.12 | 58.4 | 89.13 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| BiSeNetGFLOPs=19.912021.07 | 58.4 | 89.13 | 90.12 | — | 3,954 | — | 5,090 | 46.95 | 44.68 | 64.32 | 72.86 | 63.57 | 61.38 | 6,788 | 44.85 | 53.71 | |
| RefineNetFLOPs=44.562021.01 | 58.18 | 87.99 | 90.63 | — | 30.62 | 5,317 | 5,595 | 42.72 | 46.59 | 70.85 | 76.01 | 62.91 | 57.05 | 7,034 | 41.32 | — | |
| RefineNetGFLOPs=44.56, MParams=29.362025.12 | 58.18 | 87.99 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| RefineNetGFLOPs=44.562021.07 | 58.18 | 87.99 | 90.63 | — | 3,062 | — | 5,595 | 42.72 | 46.59 | 70.85 | 76.01 | 62.91 | 57.05 | 7,034 | 41.32 | 53.17 | |
| HardNetFLOPs=4.422021.01 | 56.19 | 90.19 | 92.87 | — | 34.62 | 4,750 | 4,240 | 49.78 | 49.19 | 62.33 | 72.93 | 68.32 | 58.14 | 6,533 | 30.9 | — | |
| HarDNetGFLOPs=4.422021.07 | 56.19 | 90.19 | 92.87 | — | 3,462 | — | 4,240 | 49.78 | 49.19 | 62.33 | 72.93 | 68.32 | 58.14 | 6,533 | 30.9 | 47.5 | |
| HRNet-w18FLOPs=4.202021.01 | 54.25 | 89.58 | 92.47 | — | 27.66 | 4,508 | 4,053 | 45.66 | 45 | 68.05 | 73.24 | 64.86 | 52.85 | 6,252 | 33.02 | — | |
| HRNet_w18GFLOPs=4.20, MParams=1.532025.12 | 54.25 | 89.58 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| HRNet w18GFLOPs=4.22021.07 | 54.25 | 89.58 | 92.47 | — | 2,766 | — | 4,053 | 45.66 | 45 | 68.05 | 73.24 | 64.86 | 52.85 | 6,252 | 33.02 | 45.08 | |
| DeepLabv3+MBv2GFLOPs=2.622021.07 | 54.16 | 88.39 | 89.95 | — | 3,179 | — | 4,618 | 41.39 | 43.42 | 61.97 | 69.48 | 61.65 | 54.89 | 6,347 | 37.36 | 48.29 | |
| FastSCNNFLOPs=1.012021.01 | 51.93 | 88.05 | 90.64 | — | 32.76 | 4,112 | 4,728 | 47.47 | 44.64 | 48.99 | 67.88 | 63.8 | 55.08 | 5,886 | 24.65 | — | |
| FastSCNNGFLOPs=1.012021.07 | 51.93 | 88.05 | 90.64 | — | 3,276 | — | 4,728 | 47.47 | 44.64 | 48.99 | 67.88 | 63.8 | 55.08 | 5,886 | 24.65 | 41.12 | |
| ContextNetFLOPs=0.872021.01 | 46.69 | 86.75 | 89.86 | — | 23.22 | 3,488 | 3,234 | 44.24 | 42.25 | 50.36 | 65.23 | 60 | 43.88 | 5,381 | 20.17 | — | |
| ContextNetGFLOPs=0.872021.07 | 46.69 | 86.75 | 89.86 | — | 2,322 | — | 3,234 | 44.24 | 42.25 | 50.36 | 65.23 | 60 | 43.88 | 5,381 | 20.17 | 34.88 | |
| LEDNetFLOPs=6.232021.01 | 46.4 | 86.07 | 88.59 | — | 28.13 | 3,672 | 3,245 | 43.77 | 38.55 | 41.51 | 64.19 | 60.05 | 42.4 | 5,312 | 27.29 | — | |
| LEDNetGFLOPs=6.232025.12 | 46.4 | 86.07 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| LEDNetGFLOPs=6.232021.07 | 46.4 | 86.07 | 88.59 | — | 2,813 | — | 3,245 | 43.77 | 38.55 | 41.51 | 64.19 | 60.05 | 42.4 | 5,312 | 27.29 | 36.72 | |
| DFANetFLOPs=1.022021.01 | 42.54 | 85.15 | 88.49 | — | 26.65 | 2,784 | 2,894 | 46.27 | 39.47 | 33.06 | 58.87 | 59.45 | 43.22 | 4,487 | 13.37 | — | |
| DFANetGFLOPs=1.022021.07 | 42.54 | 85.15 | 88.49 | — | 2,665 | — | 2,894 | 46.27 | 39.47 | 33.06 | 58.87 | 59.45 | 43.22 | 4,487 | 13.37 | 27.84 | |
| UNetFLOPs=124.552021.01 | 29.23 | 81.9 | 86.34 | — | 8.76 | 1,518 | 1,902 | 27.13 | 24.73 | 17.26 | 53.4 | 47.36 | 11.97 | 3,779 | 1.77 | — | |
| U-NetGFLOPs=124.55, MParams=13.392025.12 | 29.23 | 81.9 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| U-NetGFLOPs=124.552021.07 | 29.23 | 81.9 | 86.34 | — | 876 | — | 1,902 | 27.13 | 24.73 | 17.26 | 53.4 | 47.36 | 11.97 | 3,779 | 1.77 | 15.18 | |
| ICNetFLOPs=10.642021.01 | 23.39 | 78.23 | 83.29 | — | 2.96 | 491 | 933 | 19.24 | 15.35 | 24.11 | 44.54 | 41.49 | 7.58 | 2,747 | 3.8 | — | |
| ICNetGFLOPs=10.64, MParams=8.462025.12 | 23.39 | 78.23 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| ICNetGFLOPs=10.642021.07 | 23.39 | 78.23 | 83.29 | — | 296 | — | 933 | 19.24 | 15.35 | 24.11 | 44.54 | 41.49 | 7.58 | 2,747 | 3.8 | 4.91 | |
| DABNetFLOPs=5.182021.01 | 15.27 | 77.43 | 81.19 | — | 0 | 9 | 0 | 4.1 | 10.49 | 0 | 36.18 | 42.83 | 0 | 830 | 0 | — | |
| DABNetGFLOPs=5.182021.07 | 15.27 | 77.43 | 81.19 | — | 0 | — | 0 | 4.1 | 10.49 | 0 | 36.18 | 42.83 | 0 | 830 | 0 | 0.09 | |
| ESPNetv2FLOPs=0.832021.01 | 12.27 | 73.03 | 78.98 | — | 0 | 0 | 0 | 0 | 6.17 | 0 | 30.65 | 37.03 | 0 | 0 | 0 | — | |
| ESPNetv2GFLOPs=0.832021.07 | 12.27 | 73.03 | 78.98 | — | 0 | — | 0 | 0 | 6.17 | 0 | 30.65 | 37.03 | 0 | 0 | 0 | 0 |