Cityscapes
Models assign a class to each pixel in street scenes filmed from a vehicle. The benchmark compares predicted segments with annotated road scenes.
- Made by
- Cityscapes team
- Released
- April 2016
- Kind
- Detection and captioning
- Level
- Everyday
- Scored
- Checked automatically
- Answers
- Open-ended
- Tasks
- All public
- Measures
- Capability
- Language
- English
- Interaction
- Single turn
- Items from
- Real work and use
- Refresh
- One fixed set
- Run by
- Anyone
- In views
- Vision, Image, video and 3D generation
GEELY-ATC-SEG tops the leaderboard with 86.7%
Best score per entry from the benchmark's own leaderboard, read on November 20th, 2026. Where a model ran with several settings or harnesses, we keep its best. Names in gray aren't in our model list: fine-tunes, agent systems, products and older models. They don't count toward leaders.
- 86.7 GEELY-ATC-SEG
- 86.7 VLTSeg
- 86.5 MeiTuan-BaseModel
- 86.4 NeuroFuser
- 86.2 SAIT SeeThroughNet
- 86.1 InternImage: Exploring Large-Scale Vision Foundation Models with Deformable Convolutions
- 85.8 RefineNet
- 85.4 Hierarchical Multi-Scale Attention for Semantic Segmentation
- 85.3 DCNAS+ASPP [Mapillary Vistas]
- 85.3 mogo_semantic
- 85.3 UJS_model
- 85.2 Naive-Student (iterative semi-supervised learning with Panoptic-DeepLab)
- 85.2 Vision Transformer Adapter for Dense Predictions
- 85.1 Panoptic-DeepLab w/ SWideRNet [Mapillary Vistas + Pseudo-labels]
- 85.0 SERNet-Former_v2
- 84.8 DLA_HRNet48OCR_MSFLIP_000
- 84.6 ddl_seg
- 84.5 HRNetV2 + OCR + SegFix
- 84.5 Panoptic-DeepLab [Mapillary Vistas]
- 84.3 DCNAS+ASPP
- 84.3 Seg_UJS
- 84.2 EfficientPS [Mapillary Vistas]
- 84.1 Axial-DeepLab-XL [Mapillary Vistas]
- 84.1 Panoptic-DeepLab w/ SWideRNet [Mapillary Vistas]
- 83.7 DecoupleSegNet
- 83.7 HRNetV2 + OCR (w/ ASP)
- 83.6 DCNAS
- 83.6 Dense Prediction with Attentive Feature aggregation
- 83.5 Improving Semantic Segmentation via Video Propagation and Label Relaxation
- 83.4 Hard Pixel Mining for Depth Privileged Semantic Segmentation
- 83.3 GALD-Net
- 83.3 HRNetV2 + OCR
- 83.3 ResNeSt200
- 83.2 GGCF
- 83.2 kMaX-DeepLab [Cityscapes-fine]
- 83.2 LGE A&B Center: HANet (ResNext-101)
- 83.1 Axial-DeepLab-L [Mapillary Vistas]
- 83.1 GALD-net
- 83.0 FsaNet
- 83.0 MRFM
- 82.9 CASIA_IVA_DRANet-101_NoCoarse
- 82.8 DRN_CRL_Coarse
- 82.8 Gated-SCNN
- 82.8 SA-Gate (ResNet-101,OS=16)
- 82.7 DPC
- 82.6 HRNet + LKPP + EA loss
- 82.5 GPNet-ResNet101
- 82.4 RelationNet_Coarse
- 82.3 GFF-Net
- 82.3 SSMA
- 82.3 VCD-NoCoarse
- 82.2 DDAR
- 82.1 Auto-DeepLab-L
- 82.1 DeepLabv3+
- 82.1 ECFD
- 82.1 Fast OCNet
- 82.1 GPSNet
- 82.1 Learnable Tree Filter V2
- 82.1 LGE A&B Center: HANet (ResNet-101)
- 82.1 Margin calibration
- 82.0 DGCNet
- 82.0 FAN_NV_RVC
- 82.0 Mapillary Research: In-Place Activated BatchNorm
- 81.9 EMANet
- 81.9 SR-AIC
- 81.8 EFBNET
- 81.8 HRNetV2
- 81.8 KANet_Res101
- 81.7 EaNet-V1
- 81.6 dpcan_trainval_os16_225
- 81.6 HPNet
- 81.5 CASIA_IVA_DANet_NoCoarse
- 81.4 BFP
- 81.3 AdapNet++
- 81.3 DeepLabv3
- 81.3 SJTU_hpm
- 81.2 HRN+DCNv2_for_DOAS
- 81.2 L2-SP
- 81.2 OCNet_ResNet101_fine
- 81.2 PSPNet
- 81.1 MCDN
- 81.1 SPGNet
- 81.0 navi
- 80.9 HANet (fine-train only)
- 80.9 SANet
- 80.8 GLNet_fine
- 80.8 Learnable Tree Filter
- 80.7 MSeg1080_RVC
- 80.6 LDN-161
- 80.6 ResNet-38
- 80.6 ZLFNET
- 80.5 NetWarp
- 80.5 PGCNet_Res101_fine
- 80.5 W3_FAFM
- 80.4 Panoptic-DeepLab w/ SWideRNet [Cityscapes-fine]
- 80.3 DFN
- 80.2 EFC_sync
- 79.9 Axial-DeepLab-XL [Cityscapes-fine]
- 79.8 MIX6D_RVC
- 79.6 pixel-to-anchor contrastive learning
- 79.5 Axial-DeepLab-L [Cityscapes-fine]
- 79.5 TKCN
- 79.4 Panoptic-DeepLab [Cityscapes-fine]
- 79.4 UDSSEG_RVC
- 79.3 Knowledge-Aware
- 79.3 LDN-121
- 79.2 Roadstar.ai_CV(SFNet)
- 79.2 SegModel
- 79.1 Adaptive Affinity Field on PSPNet
- 79.0 MT-SSSR
- 78.9 PAC: Perspective-adaptive Convolutions
- 78.7 TXWone
- 78.6 PIDNet-S
- 78.5 Multitask Learning
- 78.4 ESANet RGB-D
- 78.4 Ladder DenseNet v2
- 78.2 AAF+GLR
- 78.2 depthAwareSeg_RNN_ff
- 78.1 SAC-multiple
- 77.9 LDN2_ROB
- 77.8 jntm
- 77.7 HRN
- 77.7 STDC_GAX_V2_VALTRAIN
- 77.6 ESANet RGB
- 77.3 Global-Local-Refinement
- 77.3 semantic net
- 76.8 EdgeSenseSeg
- 76.8 MSNET
- 76.7 DWGSeg-L75
- 76.5 Bilateral_attention_semantic
- 76.5 SwiftNetRN-18 ensemble
- 76.4 AFPNet
- 76.4 PIDNet
- 76.4 SwiftNet RN18 pyr sepBN MVD
- 76.0 CABiNet
- 76.0 K-net
- 75.9 FC-HarDNet-70
- 75.8 RRR-ResNet152-MultiScale
- 75.6 ESANet RGB-D (small input)
- 75.5 SwiftNetRN-18
- 75.4 PEARL
- 75.3 DF-Seg
- 75.3 SU_Net
- 75.1 IBN-PSP-SA_ROB
- 75.1 ScaleNet
- 74.7 SN_RN152pyrx8_RVC
- 74.4 EnsembleModel_Bosch
- 74.4 SwiftNetRN-18 pyramid
- 74.3 EKENet
- 74.3 Ladder DenseNet
- 74.1 FoveaNet
- 74.0 PMSDSEN
- 73.9 MaskRCNN_BOSH
- 73.7 dpfnet
- 73.4 MCANet
- 73.4 XPARSE
- 73.3 CGMANet_v1
- 73.1 NfS-Seg
- 73.0 SAITv2-light
- 72.9 ESANet RGB (small input)
- 72.8 EFRNet-13
- 72.6 Real-time FCN
- 72.5 SSNet
- 72.4 FTFNet
- 72.4 PartnerNet
- 72.4 PBRNet
- 72.4 SDBNetV2
- 72.2 ALV303
- 71.9 MNet_MPRG
- 71.8 FRRN
- 71.8 LRR-4x
- 71.8 m-TCFs
- 71.6 Adelaide_context
- 71.5 FasterSeg
- 71.4 DSANet: Dilated Spatial Attention for Real-time Semantic Segmentation in Urban Street Scenes
- 71.3 LDFNet
- 71.3 YOLO V5s with Segmentation Head
- 71.2 ML-CRNN
- 71.1 Deep Layer Cascade (LC)
- 71.1 IDWRSeg_Deform
- 71.0 BFNet
- 71.0 Edgenet
- 70.8 LiteSeg-Darknet19
- 70.8 SDBNet
- 70.7 MobileNetV2Plus
- 70.7 RFMobileNetV2Plus
- 70.6 AHiSS_ROB
- 70.6 SFRSeg
- 70.4 DeepLabv2-CRF
- 70.4 GUNet
- 70.3 ShuffleNet v2 + DPC
- 70.2 F2MF-short
- 70.1 AttaNet_light
- 70.1 CFPNet
- 70.0 SAITv2
- 69.8 ERINet
- 69.8 EVANet
- 69.8 GridNet
- 69.7 ERFNet (pretrained)
- 69.4 FSFFNet
- 69.4 SPSSN
- 69.3 DSNet
- 69.1 FSFNet
- 69.1 PL-Seg
- 68.9 Spatial Sampling Net
- 68.4 FarSee-Net
- 68.3 M2FANet
- 68.3 RPNet
- 68.3 SCMNet coarse
- 68.2 firenet
- 68.0 ERFNet (from scratch)
- 68.0 Fast-SCNN
- 68.0 MiniNet-v2-pretrained
- 67.9 SCMNet
- 67.8 LiteSeg-Mobilenet
- 67.8 MiniNet-v2-coarse
- 67.6 ERSNet-coarse
- 67.4 ERINet-v2
- 67.4 NVSegNet
- 67.4 RGB-D FCN
- 67.3 EDANet
- 67.3 pruned & dilated inception-resnet-v2 (PD-IR2)
- 67.1 Dilation10
- 67.0 seamseg_rvcsubset
- 66.8 DPN
- 66.4 VENUS_ROB
- 66.3 Scale invariant CNN + CRF
- 66.2 ESPNetv2
- 66.1 ContextNet
- 65.9 Le_Selfdriving_VGG
- 65.8 DeepLabv3+ decoder
- 65.8 FreeNet
- 65.8 Hybrid Model
- 65.3 FCN 8s
- 65.2 LiteSeg-Shufflenet
- 64.8 C3Net
- 64.8 CGNet
- 64.8 DeepLab LargeFOV StrongWeak
- 64.8 ndnet
- 64.5 ADSCNet
- 64.5 DeepLab+DynamicCRF
- 64.5 RMNet
- 64.3 Pixel-level Encoding for Instance Segmentation
- 64.1 FANet
- 64.0 ThunderNet
- 63.9 Mobilenetv3-small-backbone real-time segmentation
- 63.8 AdapNetv2_ROB
- 63.3 CLRCNet
- 63.2 EffPS_b1bs4_RVC
- 63.1 DeepLab LargeFOV Strong
- 63.1 ENet with the Lovász-Softmax loss
- 63.0 BiSeNetV2-CityLabeler
- 63.0 GoogLeNet FCN
- 62.8 Fast-SCNN (Half-resolution)
- 62.7 VlocNet++_ROB
- 62.5 CRFasRNN
- 61.5 SkipNet-MobileNet
- 61.0 LBPSS
- 60.3 ESPNet
- 59.8 SQ
- 59.6 GoogLeNetV1_ROB
- 59.3 MultiBoost
- 59.1 F2MF-mid
- 58.9 HiSS_ROB
- 58.3 ENet
- 58.3 ShuffleSeg
- 57.4 FCN8s-QunjieYu
- 57.0 Segnet basic
- 56.5 APMoE_seg_ROB
- 56.2 BiSeNet-V10
- 56.1 Segnet extended
- 55.4 BatMAN_ROB
- 51.9 Fast-SCNN (Quarter-resolution)
- 51.3 Deform_ResNet_Balanced
- 48.0 UFONet (half-resolution)
- 40.7 MiniNet
- 30.4 FCN101_ROB
- 30.0 RFLR
- 1 GEELY-ATC-SEG 86.7%
- 2 VLTSeg Submission 20118 86.7%
- 3 MeiTuan-BaseModel 86.5%
- 4 NeuroFuser 86.4%
- 5 SAIT SeeThroughNet 86.2%
- 6 InternImage: Exploring Large-Scale Vision Foundation Models with Deformable Convolutions 86.1%
- 7 RefineNet Submission 9210 85.8%
- 8 Hierarchical Multi-Scale Attention for Semantic Segmentation 85.4%
- 9 DCNAS+ASPP [Mapillary Vistas] 85.3%
- 10 mogo_semantic 85.3%
- 11 UJS_model 85.3%
- 12 Naive-Student (iterative semi-supervised learning with Panoptic-DeepLab) 85.2%
- 13 Vision Transformer Adapter for Dense Predictions 85.2%
- 14 Panoptic-DeepLab w/ SWideRNet [Mapillary Vistas + Pseudo-labels] 85.1%
- 15 SERNet-Former_v2 85%
- 16 DLA_HRNet48OCR_MSFLIP_000 84.8%
- 17 ddl_seg 84.6%
- 18 HRNetV2 + OCR + SegFix 84.5%
- 19 Panoptic-DeepLab [Mapillary Vistas] 84.5%
- 20 DCNAS+ASPP 84.3%
- 21 Seg_UJS 84.3%
- 22 EfficientPS [Mapillary Vistas] 84.2%
- 23 Axial-DeepLab-XL [Mapillary Vistas] 84.1%
- 24 Panoptic-DeepLab w/ SWideRNet [Mapillary Vistas] 84.1%
- 25 DecoupleSegNet 83.7%
- 26 HRNetV2 + OCR (w/ ASP) 83.7%
- 27 DCNAS 83.6%
- 28 Dense Prediction with Attentive Feature aggregation 83.6%
- 29 Improving Semantic Segmentation via Video Propagation and Label Relaxation 83.5%
- 30 Hard Pixel Mining for Depth Privileged Semantic Segmentation 83.4%
- 31 GALD-Net 83.3%
- 32 HRNetV2 + OCR 83.3%
- 33 ResNeSt200 83.3%
- 34 GGCF 83.2%
- 35 kMaX-DeepLab [Cityscapes-fine] 83.2%
- 36 LGE A&B Center: HANet (ResNext-101) 83.2%
- 37 Axial-DeepLab-L [Mapillary Vistas] 83.1%
- 38 GALD-net 83.1%
- 39 FsaNet 83%
- 40 MRFM 83%
- 41 CASIA_IVA_DRANet-101_NoCoarse 82.9%
- 42 DRN_CRL_Coarse 82.8%
- 43 Gated-SCNN 82.8%
- 44 SA-Gate (ResNet-101,OS=16) 82.8%
- 45 DPC 82.7%
- 46 HRNet + LKPP + EA loss 82.6%
- 47 GPNet-ResNet101 82.5%
- 48 RelationNet_Coarse 82.4%
- 49 GFF-Net 82.3%
- 50 SSMA 82.3%
- 51 VCD-NoCoarse 82.3%
- 52 DDAR 82.2%
- 53 Auto-DeepLab-L 82.1%
- 54 DeepLabv3+ 82.1%
- 55 ECFD 82.1%
- 56 Fast OCNet 82.1%
- 57 GPSNet 82.1%
- 58 Learnable Tree Filter V2 82.1%
- 59 LGE A&B Center: HANet (ResNet-101) 82.1%
- 60 Margin calibration 82.1%
- 61 DGCNet 82%
- 62 FAN_NV_RVC 82%
- 63 Mapillary Research: In-Place Activated BatchNorm 82%
- 64 EMANet 81.9%
- 65 SR-AIC 81.9%
- 66 EFBNET 81.8%
- 67 HRNetV2 81.8%
- 68 KANet_Res101 81.8%
- 69 EaNet-V1 81.7%
- 70 dpcan_trainval_os16_225 81.6%
- 71 HPNet 81.6%
- 72 CASIA_IVA_DANet_NoCoarse 81.5%
- 73 BFP 81.4%
- 74 AdapNet++ 81.3%
- 75 DeepLabv3 81.3%
- 76 SJTU_hpm 81.3%
- 77 HRN+DCNv2_for_DOAS 81.2%
- 78 L2-SP 81.2%
- 79 OCNet_ResNet101_fine 81.2%
- 80 PSPNet 81.2%
- 81 MCDN 81.1%
- 82 SPGNet 81.1%
- 83 navi 81%
- 84 HANet (fine-train only) 80.9%
- 85 SANet 80.9%
- 86 GLNet_fine 80.8%
- 87 Learnable Tree Filter 80.8%
- 88 MSeg1080_RVC 80.7%
- 89 LDN-161 80.6%
- 90 ResNet-38 Submission 361 80.6%
- 91 ZLFNET 80.6%
- 92 NetWarp 80.5%
- 93 PGCNet_Res101_fine 80.5%
- 94 W3_FAFM 80.5%
- 95 Panoptic-DeepLab w/ SWideRNet [Cityscapes-fine] 80.4%
- 96 DFN 80.3%
- 97 EFC_sync 80.2%
- 98 Axial-DeepLab-XL [Cityscapes-fine] 79.9%
- 99 MIX6D_RVC 79.8%
- 100 pixel-to-anchor contrastive learning 79.6%
- 101 Axial-DeepLab-L [Cityscapes-fine] 79.5%
- 102 TKCN 79.5%
- 103 Panoptic-DeepLab [Cityscapes-fine] 79.4%
- 104 UDSSEG_RVC 79.4%
- 105 Knowledge-Aware 79.3%
- 106 LDN-121 79.3%
- 107 Roadstar.ai_CV(SFNet) 79.2%
- 108 SegModel Submission 335 79.2%
- 109 Adaptive Affinity Field on PSPNet 79.1%
- 110 MT-SSSR 79%
- 111 PAC: Perspective-adaptive Convolutions 78.9%
- 112 TXWone 78.7%
- 113 PIDNet-S 78.6%
- 114 Multitask Learning 78.5%
- 115 ESANet RGB-D 78.4%
- 116 Ladder DenseNet v2 78.4%
- 117 AAF+GLR 78.2%
- 118 depthAwareSeg_RNN_ff 78.2%
- 119 SAC-multiple 78.1%
- 120 LDN2_ROB 77.9%
- 121 jntm 77.8%
- 122 HRN 77.7%
- 123 STDC_GAX_V2_VALTRAIN 77.7%
- 124 ESANet RGB 77.6%
- 125 Global-Local-Refinement 77.3%
- 126 semantic net 77.3%
- 127 EdgeSenseSeg 76.8%
- 128 MSNET 76.8%
- 129 DWGSeg-L75 76.7%
- 130 Bilateral_attention_semantic 76.5%
- 131 SwiftNetRN-18 ensemble 76.5%
- 132 AFPNet 76.4%
- 133 PIDNet 76.4%
- 134 SwiftNet RN18 pyr sepBN MVD 76.4%
- 135 CABiNet 76%
- 136 K-net 76%
- 137 FC-HarDNet-70 75.9%
- 138 RRR-ResNet152-MultiScale 75.8%
- 139 ESANet RGB-D (small input) 75.6%
- 140 SwiftNetRN-18 75.5%
- 141 PEARL 75.4%
- 142 DF-Seg Submission 2757 75.3%
- 143 SU_Net 75.3%
- 144 IBN-PSP-SA_ROB 75.1%
- 145 ScaleNet 75.1%
- 146 SN_RN152pyrx8_RVC 74.7%
- 147 EnsembleModel_Bosch 74.4%
- 148 SwiftNetRN-18 pyramid 74.4%
- 149 EKENet 74.3%
- 150 Ladder DenseNet 74.3%
- 151 FoveaNet 74.1%
- 152 PMSDSEN 74%
- 153 MaskRCNN_BOSH 73.9%
- 154 dpfnet 73.7%
- 155 MCANet 73.4%
- 156 XPARSE 73.4%
- 157 CGMANet_v1 73.3%
- 158 NfS-Seg 73.1%
- 159 SAITv2-light 73%
- 160 ESANet RGB (small input) 72.9%
- 161 EFRNet-13 72.8%
- 162 Real-time FCN 72.6%
- 163 SSNet 72.5%
- 164 FTFNet 72.4%
- 165 PartnerNet 72.4%
- 166 PBRNet 72.4%
- 167 SDBNetV2 72.4%
- 168 ALV303 72.2%
- 169 MNet_MPRG 71.9%
- 170 FRRN 71.8%
- 171 LRR-4x Submission 198 71.8%
- 172 m-TCFs 71.8%
- 173 Adelaide_context 71.6%
- 174 FasterSeg 71.5%
- 175 DSANet: Dilated Spatial Attention for Real-time Semantic Segmentation in Urban Street Scenes 71.4%
- 176 LDFNet 71.3%
- 177 YOLO V5s with Segmentation Head 71.3%
- 178 ML-CRNN 71.2%
- 179 Deep Layer Cascade (LC) 71.1%
- 180 IDWRSeg_Deform 71.1%
- 181 BFNet 71%
- 182 Edgenet 71%
- 183 LiteSeg-Darknet19 70.8%
- 184 SDBNet 70.8%
- 185 MobileNetV2Plus 70.7%
- 186 RFMobileNetV2Plus 70.7%
- 187 AHiSS_ROB 70.6%
- 188 SFRSeg 70.6%
- 189 DeepLabv2-CRF 70.4%
- 190 GUNet 70.4%
- 191 ShuffleNet v2 + DPC 70.3%
- 192 F2MF-short 70.2%
- 193 AttaNet_light 70.1%
- 194 CFPNet 70.1%
- 195 SAITv2 70%
- 196 ERINet 69.8%
- 197 EVANet 69.8%
- 198 GridNet Submission 653 69.8%
- 199 ERFNet (pretrained) 69.7%
- 200 FSFFNet 69.4%
- 201 SPSSN 69.4%
- 202 DSNet Submission 2621 69.3%
- 203 FSFNet 69.1%
- 204 PL-Seg 69.1%
- 205 Spatial Sampling Net 68.9%
- 206 FarSee-Net 68.4%
- 207 M2FANet 68.3%
- 208 RPNet 68.3%
- 209 SCMNet coarse 68.3%
- 210 firenet 68.2%
- 211 ERFNet (from scratch) 68%
- 212 Fast-SCNN 68%
- 213 MiniNet-v2-pretrained 68%
- 214 SCMNet 67.9%
- 215 LiteSeg-Mobilenet 67.8%
- 216 MiniNet-v2-coarse 67.8%
- 217 ERSNet-coarse 67.6%
- 218 ERINet-v2 67.4%
- 219 NVSegNet 67.4%
- 220 RGB-D FCN 67.4%
- 221 EDANet 67.3%
- 222 pruned & dilated inception-resnet-v2 (PD-IR2) 67.3%
- 223 Dilation10 67.1%
- 224 seamseg_rvcsubset 67%
- 225 DPN Submission 82 66.8%
- 226 VENUS_ROB 66.4%
- 227 Scale invariant CNN + CRF 66.3%
- 228 ESPNetv2 66.2%
- 229 ContextNet 66.1%
- 230 Le_Selfdriving_VGG 65.9%
- 231 DeepLabv3+ decoder 65.8%
- 232 FreeNet 65.8%
- 233 Hybrid Model 65.8%
- 234 FCN 8s 65.3%
- 235 LiteSeg-Shufflenet 65.2%
- 236 C3Net 64.8%
- 237 CGNet 64.8%
- 238 DeepLab LargeFOV StrongWeak 64.8%
- 239 ndnet 64.8%
- 240 ADSCNet 64.5%
- 241 DeepLab+DynamicCRF 64.5%
- 242 RMNet 64.5%
- 243 Pixel-level Encoding for Instance Segmentation 64.3%
- 244 FANet 64.1%
- 245 ThunderNet 64%
- 246 Mobilenetv3-small-backbone real-time segmentation 63.9%
- 247 AdapNetv2_ROB 63.8%
- 248 CLRCNet 63.3%
- 249 EffPS_b1bs4_RVC 63.2%
- 250 DeepLab LargeFOV Strong 63.1%
- 251 ENet with the Lovász-Softmax loss 63.1%
- 252 BiSeNetV2-CityLabeler 63%
- 253 GoogLeNet FCN 63%
- 254 Fast-SCNN (Half-resolution) 62.8%
- 255 VlocNet++_ROB 62.7%
- 256 CRFasRNN 62.5%
- 257 SkipNet-MobileNet 61.5%
- 258 LBPSS 61%
- 259 ESPNet 60.3%
- 260 SQ 59.8%
- 261 GoogLeNetV1_ROB 59.6%
- 262 MultiBoost 59.3%
- 263 F2MF-mid 59.1%
- 264 HiSS_ROB 58.9%
- 265 ENet 58.3%
- 266 ShuffleSeg 58.3%
- 267 FCN8s-QunjieYu 57.4%
- 268 Segnet basic 57%
- 269 APMoE_seg_ROB 56.5%
- 270 BiSeNet-V10 56.2%
- 271 Segnet extended 56.1%
- 272 BatMAN_ROB 55.4%
- 273 Fast-SCNN (Quarter-resolution) 51.9%
- 274 Deform_ResNet_Balanced 51.3%
- 275 UFONet (half-resolution) 48%
- 276 MiniNet 40.7%
- 277 FCN101_ROB 30.4%
- 278 RFLR 30%
Source: Official Cityscapes leaderboard
| # | Entry | Score | Setting | Submitted by | Date | Link | Source |
|---|---|---|---|---|---|---|---|
| 1 | GEELY-ATC-SEG | 86.7% | cityscapes-dataset.com | Official leaderboard | |||
| 2 | VLTSeg | 86.7% | Submission 20118 | HW4 | cityscapes-dataset.com | Official leaderboard | |
| 3 | MeiTuan-BaseModel | 86.5% | cityscapes-dataset.com | Official leaderboard | |||
| 4 | NeuroFuser | 86.4% | cityscapes-dataset.com | Official leaderboard | |||
| 5 | SAIT SeeThroughNet | 86.2% | cityscapes-dataset.com | Official leaderboard | |||
| 6 | InternImage: Exploring Large-Scale Vision Foundation Models with Deformable Convolutions | 86.1% | cityscapes-dataset.com | Official leaderboard | |||
| 7 | RefineNet | 85.8% | Submission 9210 | DAHUA-ARI | cityscapes-dataset.com | Official leaderboard | |
| 8 | Hierarchical Multi-Scale Attention for Semantic Segmentation | 85.4% | cityscapes-dataset.com | Official leaderboard | |||
| 9 | DCNAS+ASPP [Mapillary Vistas] | 85.3% | cityscapes-dataset.com | Official leaderboard | |||
| 10 | mogo_semantic | 85.3% | cityscapes-dataset.com | Official leaderboard | |||
| 11 | UJS_model | 85.3% | cityscapes-dataset.com | Official leaderboard | |||
| 12 | Naive-Student (iterative semi-supervised learning with Panoptic-DeepLab) | 85.2% | cityscapes-dataset.com | Official leaderboard | |||
| 13 | Vision Transformer Adapter for Dense Predictions | 85.2% | cityscapes-dataset.com | Official leaderboard | |||
| 14 | Panoptic-DeepLab w/ SWideRNet [Mapillary Vistas + Pseudo-labels] | 85.1% | cityscapes-dataset.com | Official leaderboard | |||
| 15 | SERNet-Former_v2 | 85% | cityscapes-dataset.com | Official leaderboard | |||
| 16 | DLA_HRNet48OCR_MSFLIP_000 | 84.8% | cityscapes-dataset.com | Official leaderboard | |||
| 17 | ddl_seg | 84.6% | cityscapes-dataset.com | Official leaderboard | |||
| 18 | HRNetV2 + OCR + SegFix | 84.5% | cityscapes-dataset.com | Official leaderboard | |||
| 19 | Panoptic-DeepLab [Mapillary Vistas] | 84.5% | cityscapes-dataset.com | Official leaderboard | |||
| 20 | DCNAS+ASPP | 84.3% | cityscapes-dataset.com | Official leaderboard | |||
| 21 | Seg_UJS | 84.3% | cityscapes-dataset.com | Official leaderboard | |||
| 22 | EfficientPS [Mapillary Vistas] | 84.2% | cityscapes-dataset.com | Official leaderboard | |||
| 23 | Axial-DeepLab-XL [Mapillary Vistas] | 84.1% | cityscapes-dataset.com | Official leaderboard | |||
| 24 | Panoptic-DeepLab w/ SWideRNet [Mapillary Vistas] | 84.1% | cityscapes-dataset.com | Official leaderboard | |||
| 25 | DecoupleSegNet | 83.7% | cityscapes-dataset.com | Official leaderboard | |||
| 26 | HRNetV2 + OCR (w/ ASP) | 83.7% | cityscapes-dataset.com | Official leaderboard | |||
| 27 | DCNAS | 83.6% | cityscapes-dataset.com | Official leaderboard | |||
| 28 | Dense Prediction with Attentive Feature aggregation | 83.6% | cityscapes-dataset.com | Official leaderboard | |||
| 29 | Improving Semantic Segmentation via Video Propagation and Label Relaxation | 83.5% | cityscapes-dataset.com | Official leaderboard | |||
| 30 | Hard Pixel Mining for Depth Privileged Semantic Segmentation | 83.4% | cityscapes-dataset.com | Official leaderboard | |||
| 31 | GALD-Net | 83.3% | cityscapes-dataset.com | Official leaderboard | |||
| 32 | HRNetV2 + OCR | 83.3% | cityscapes-dataset.com | Official leaderboard | |||
| 33 | ResNeSt200 | 83.3% | cityscapes-dataset.com | Official leaderboard | |||
| 34 | GGCF | 83.2% | cityscapes-dataset.com | Official leaderboard | |||
| 35 | kMaX-DeepLab [Cityscapes-fine] | 83.2% | cityscapes-dataset.com | Official leaderboard | |||
| 36 | LGE A&B Center: HANet (ResNext-101) | 83.2% | cityscapes-dataset.com | Official leaderboard | |||
| 37 | Axial-DeepLab-L [Mapillary Vistas] | 83.1% | cityscapes-dataset.com | Official leaderboard | |||
| 38 | GALD-net | 83.1% | cityscapes-dataset.com | Official leaderboard | |||
| 39 | FsaNet | 83% | cityscapes-dataset.com | Official leaderboard | |||
| 40 | MRFM | 83% | cityscapes-dataset.com | Official leaderboard | |||
| 41 | CASIA_IVA_DRANet-101_NoCoarse | 82.9% | cityscapes-dataset.com | Official leaderboard | |||
| 42 | DRN_CRL_Coarse | 82.8% | cityscapes-dataset.com | Official leaderboard | |||
| 43 | Gated-SCNN | 82.8% | cityscapes-dataset.com | Official leaderboard | |||
| 44 | SA-Gate (ResNet-101,OS=16) | 82.8% | cityscapes-dataset.com | Official leaderboard | |||
| 45 | DPC | 82.7% | cityscapes-dataset.com | Official leaderboard | |||
| 46 | HRNet + LKPP + EA loss | 82.6% | cityscapes-dataset.com | Official leaderboard | |||
| 47 | GPNet-ResNet101 | 82.5% | cityscapes-dataset.com | Official leaderboard | |||
| 48 | RelationNet_Coarse | 82.4% | cityscapes-dataset.com | Official leaderboard | |||
| 49 | GFF-Net | 82.3% | cityscapes-dataset.com | Official leaderboard | |||
| 50 | SSMA | 82.3% | cityscapes-dataset.com | Official leaderboard | |||
| 51 | VCD-NoCoarse | 82.3% | cityscapes-dataset.com | Official leaderboard | |||
| 52 | DDAR | 82.2% | cityscapes-dataset.com | Official leaderboard | |||
| 53 | Auto-DeepLab-L | 82.1% | cityscapes-dataset.com | Official leaderboard | |||
| 54 | DeepLabv3+ | 82.1% | cityscapes-dataset.com | Official leaderboard | |||
| 55 | ECFD | 82.1% | cityscapes-dataset.com | Official leaderboard | |||
| 56 | Fast OCNet | 82.1% | cityscapes-dataset.com | Official leaderboard | |||
| 57 | GPSNet | 82.1% | cityscapes-dataset.com | Official leaderboard | |||
| 58 | Learnable Tree Filter V2 | 82.1% | cityscapes-dataset.com | Official leaderboard | |||
| 59 | LGE A&B Center: HANet (ResNet-101) | 82.1% | cityscapes-dataset.com | Official leaderboard | |||
| 60 | Margin calibration | 82.1% | cityscapes-dataset.com | Official leaderboard | |||
| 61 | DGCNet | 82% | cityscapes-dataset.com | Official leaderboard | |||
| 62 | FAN_NV_RVC | 82% | cityscapes-dataset.com | Official leaderboard | |||
| 63 | Mapillary Research: In-Place Activated BatchNorm | 82% | cityscapes-dataset.com | Official leaderboard | |||
| 64 | EMANet | 81.9% | cityscapes-dataset.com | Official leaderboard | |||
| 65 | SR-AIC | 81.9% | cityscapes-dataset.com | Official leaderboard | |||
| 66 | EFBNET | 81.8% | cityscapes-dataset.com | Official leaderboard | |||
| 67 | HRNetV2 | 81.8% | cityscapes-dataset.com | Official leaderboard | |||
| 68 | KANet_Res101 | 81.8% | cityscapes-dataset.com | Official leaderboard | |||
| 69 | EaNet-V1 | 81.7% | cityscapes-dataset.com | Official leaderboard | |||
| 70 | dpcan_trainval_os16_225 | 81.6% | cityscapes-dataset.com | Official leaderboard | |||
| 71 | HPNet | 81.6% | cityscapes-dataset.com | Official leaderboard | |||
| 72 | CASIA_IVA_DANet_NoCoarse | 81.5% | cityscapes-dataset.com | Official leaderboard | |||
| 73 | BFP | 81.4% | cityscapes-dataset.com | Official leaderboard | |||
| 74 | AdapNet++ | 81.3% | cityscapes-dataset.com | Official leaderboard | |||
| 75 | DeepLabv3 | 81.3% | cityscapes-dataset.com | Official leaderboard | |||
| 76 | SJTU_hpm | 81.3% | cityscapes-dataset.com | Official leaderboard | |||
| 77 | HRN+DCNv2_for_DOAS | 81.2% | cityscapes-dataset.com | Official leaderboard | |||
| 78 | L2-SP | 81.2% | cityscapes-dataset.com | Official leaderboard | |||
| 79 | OCNet_ResNet101_fine | 81.2% | cityscapes-dataset.com | Official leaderboard | |||
| 80 | PSPNet | 81.2% | cityscapes-dataset.com | Official leaderboard | |||
| 81 | MCDN | 81.1% | cityscapes-dataset.com | Official leaderboard | |||
| 82 | SPGNet | 81.1% | cityscapes-dataset.com | Official leaderboard | |||
| 83 | navi | 81% | cityscapes-dataset.com | Official leaderboard | |||
| 84 | HANet (fine-train only) | 80.9% | cityscapes-dataset.com | Official leaderboard | |||
| 85 | SANet | 80.9% | cityscapes-dataset.com | Official leaderboard | |||
| 86 | GLNet_fine | 80.8% | cityscapes-dataset.com | Official leaderboard | |||
| 87 | Learnable Tree Filter | 80.8% | cityscapes-dataset.com | Official leaderboard | |||
| 88 | MSeg1080_RVC | 80.7% | cityscapes-dataset.com | Official leaderboard | |||
| 89 | LDN-161 | 80.6% | cityscapes-dataset.com | Official leaderboard | |||
| 90 | ResNet-38 | 80.6% | Submission 361 | cityscapes-dataset.com | Official leaderboard | ||
| 91 | ZLFNET | 80.6% | cityscapes-dataset.com | Official leaderboard | |||
| 92 | NetWarp | 80.5% | cityscapes-dataset.com | Official leaderboard | |||
| 93 | PGCNet_Res101_fine | 80.5% | cityscapes-dataset.com | Official leaderboard | |||
| 94 | W3_FAFM | 80.5% | cityscapes-dataset.com | Official leaderboard | |||
| 95 | Panoptic-DeepLab w/ SWideRNet [Cityscapes-fine] | 80.4% | cityscapes-dataset.com | Official leaderboard | |||
| 96 | DFN | 80.3% | cityscapes-dataset.com | Official leaderboard | |||
| 97 | EFC_sync | 80.2% | cityscapes-dataset.com | Official leaderboard | |||
| 98 | Axial-DeepLab-XL [Cityscapes-fine] | 79.9% | cityscapes-dataset.com | Official leaderboard | |||
| 99 | MIX6D_RVC | 79.8% | cityscapes-dataset.com | Official leaderboard | |||
| 100 | pixel-to-anchor contrastive learning | 79.6% | cityscapes-dataset.com | Official leaderboard | |||
| 101 | Axial-DeepLab-L [Cityscapes-fine] | 79.5% | cityscapes-dataset.com | Official leaderboard | |||
| 102 | TKCN | 79.5% | cityscapes-dataset.com | Official leaderboard | |||
| 103 | Panoptic-DeepLab [Cityscapes-fine] | 79.4% | cityscapes-dataset.com | Official leaderboard | |||
| 104 | UDSSEG_RVC | 79.4% | cityscapes-dataset.com | Official leaderboard | |||
| 105 | Knowledge-Aware | 79.3% | cityscapes-dataset.com | Official leaderboard | |||
| 106 | LDN-121 | 79.3% | cityscapes-dataset.com | Official leaderboard | |||
| 107 | Roadstar.ai_CV(SFNet) | 79.2% | cityscapes-dataset.com | Official leaderboard | |||
| 108 | SegModel | 79.2% | Submission 335 | cityscapes-dataset.com | Official leaderboard | ||
| 109 | Adaptive Affinity Field on PSPNet | 79.1% | cityscapes-dataset.com | Official leaderboard | |||
| 110 | MT-SSSR | 79% | cityscapes-dataset.com | Official leaderboard | |||
| 111 | PAC: Perspective-adaptive Convolutions | 78.9% | cityscapes-dataset.com | Official leaderboard | |||
| 112 | TXWone | 78.7% | cityscapes-dataset.com | Official leaderboard | |||
| 113 | PIDNet-S | 78.6% | cityscapes-dataset.com | Official leaderboard | |||
| 114 | Multitask Learning | 78.5% | cityscapes-dataset.com | Official leaderboard | |||
| 115 | ESANet RGB-D | 78.4% | cityscapes-dataset.com | Official leaderboard | |||
| 116 | Ladder DenseNet v2 | 78.4% | cityscapes-dataset.com | Official leaderboard | |||
| 117 | AAF+GLR | 78.2% | cityscapes-dataset.com | Official leaderboard | |||
| 118 | depthAwareSeg_RNN_ff | 78.2% | cityscapes-dataset.com | Official leaderboard | |||
| 119 | SAC-multiple | 78.1% | cityscapes-dataset.com | Official leaderboard | |||
| 120 | LDN2_ROB | 77.9% | cityscapes-dataset.com | Official leaderboard | |||
| 121 | jntm | 77.8% | cityscapes-dataset.com | Official leaderboard | |||
| 122 | HRN | 77.7% | cityscapes-dataset.com | Official leaderboard | |||
| 123 | STDC_GAX_V2_VALTRAIN | 77.7% | cityscapes-dataset.com | Official leaderboard | |||
| 124 | ESANet RGB | 77.6% | cityscapes-dataset.com | Official leaderboard | |||
| 125 | Global-Local-Refinement | 77.3% | cityscapes-dataset.com | Official leaderboard | |||
| 126 | semantic net | 77.3% | cityscapes-dataset.com | Official leaderboard | |||
| 127 | EdgeSenseSeg | 76.8% | cityscapes-dataset.com | Official leaderboard | |||
| 128 | MSNET | 76.8% | cityscapes-dataset.com | Official leaderboard | |||
| 129 | DWGSeg-L75 | 76.7% | cityscapes-dataset.com | Official leaderboard | |||
| 130 | Bilateral_attention_semantic | 76.5% | cityscapes-dataset.com | Official leaderboard | |||
| 131 | SwiftNetRN-18 ensemble | 76.5% | cityscapes-dataset.com | Official leaderboard | |||
| 132 | AFPNet | 76.4% | cityscapes-dataset.com | Official leaderboard | |||
| 133 | PIDNet | 76.4% | cityscapes-dataset.com | Official leaderboard | |||
| 134 | SwiftNet RN18 pyr sepBN MVD | 76.4% | cityscapes-dataset.com | Official leaderboard | |||
| 135 | CABiNet | 76% | cityscapes-dataset.com | Official leaderboard | |||
| 136 | K-net | 76% | cityscapes-dataset.com | Official leaderboard | |||
| 137 | FC-HarDNet-70 | 75.9% | cityscapes-dataset.com | Official leaderboard | |||
| 138 | RRR-ResNet152-MultiScale | 75.8% | cityscapes-dataset.com | Official leaderboard | |||
| 139 | ESANet RGB-D (small input) | 75.6% | cityscapes-dataset.com | Official leaderboard | |||
| 140 | SwiftNetRN-18 | 75.5% | cityscapes-dataset.com | Official leaderboard | |||
| 141 | PEARL | 75.4% | cityscapes-dataset.com | Official leaderboard | |||
| 142 | DF-Seg | 75.3% | Submission 2757 | cityscapes-dataset.com | Official leaderboard | ||
| 143 | SU_Net | 75.3% | cityscapes-dataset.com | Official leaderboard | |||
| 144 | IBN-PSP-SA_ROB | 75.1% | cityscapes-dataset.com | Official leaderboard | |||
| 145 | ScaleNet | 75.1% | cityscapes-dataset.com | Official leaderboard | |||
| 146 | SN_RN152pyrx8_RVC | 74.7% | cityscapes-dataset.com | Official leaderboard | |||
| 147 | EnsembleModel_Bosch | 74.4% | cityscapes-dataset.com | Official leaderboard | |||
| 148 | SwiftNetRN-18 pyramid | 74.4% | cityscapes-dataset.com | Official leaderboard | |||
| 149 | EKENet | 74.3% | cityscapes-dataset.com | Official leaderboard | |||
| 150 | Ladder DenseNet | 74.3% | cityscapes-dataset.com | Official leaderboard | |||
| 151 | FoveaNet | 74.1% | cityscapes-dataset.com | Official leaderboard | |||
| 152 | PMSDSEN | 74% | cityscapes-dataset.com | Official leaderboard | |||
| 153 | MaskRCNN_BOSH | 73.9% | cityscapes-dataset.com | Official leaderboard | |||
| 154 | dpfnet | 73.7% | cityscapes-dataset.com | Official leaderboard | |||
| 155 | MCANet | 73.4% | cityscapes-dataset.com | Official leaderboard | |||
| 156 | XPARSE | 73.4% | cityscapes-dataset.com | Official leaderboard | |||
| 157 | CGMANet_v1 | 73.3% | cityscapes-dataset.com | Official leaderboard | |||
| 158 | NfS-Seg | 73.1% | cityscapes-dataset.com | Official leaderboard | |||
| 159 | SAITv2-light | 73% | cityscapes-dataset.com | Official leaderboard | |||
| 160 | ESANet RGB (small input) | 72.9% | cityscapes-dataset.com | Official leaderboard | |||
| 161 | EFRNet-13 | 72.8% | cityscapes-dataset.com | Official leaderboard | |||
| 162 | Real-time FCN | 72.6% | cityscapes-dataset.com | Official leaderboard | |||
| 163 | SSNet | 72.5% | cityscapes-dataset.com | Official leaderboard | |||
| 164 | FTFNet | 72.4% | cityscapes-dataset.com | Official leaderboard | |||
| 165 | PartnerNet | 72.4% | cityscapes-dataset.com | Official leaderboard | |||
| 166 | PBRNet | 72.4% | cityscapes-dataset.com | Official leaderboard | |||
| 167 | SDBNetV2 | 72.4% | cityscapes-dataset.com | Official leaderboard | |||
| 168 | ALV303 | 72.2% | cityscapes-dataset.com | Official leaderboard | |||
| 169 | MNet_MPRG | 71.9% | cityscapes-dataset.com | Official leaderboard | |||
| 170 | FRRN | 71.8% | cityscapes-dataset.com | Official leaderboard | |||
| 171 | LRR-4x | 71.8% | Submission 198 | cityscapes-dataset.com | Official leaderboard | ||
| 172 | m-TCFs | 71.8% | cityscapes-dataset.com | Official leaderboard | |||
| 173 | Adelaide_context | 71.6% | cityscapes-dataset.com | Official leaderboard | |||
| 174 | FasterSeg | 71.5% | cityscapes-dataset.com | Official leaderboard | |||
| 175 | DSANet: Dilated Spatial Attention for Real-time Semantic Segmentation in Urban Street Scenes | 71.4% | cityscapes-dataset.com | Official leaderboard | |||
| 176 | LDFNet | 71.3% | cityscapes-dataset.com | Official leaderboard | |||
| 177 | YOLO V5s with Segmentation Head | 71.3% | cityscapes-dataset.com | Official leaderboard | |||
| 178 | ML-CRNN | 71.2% | cityscapes-dataset.com | Official leaderboard | |||
| 179 | Deep Layer Cascade (LC) | 71.1% | cityscapes-dataset.com | Official leaderboard | |||
| 180 | IDWRSeg_Deform | 71.1% | cityscapes-dataset.com | Official leaderboard | |||
| 181 | BFNet | 71% | cityscapes-dataset.com | Official leaderboard | |||
| 182 | Edgenet | 71% | cityscapes-dataset.com | Official leaderboard | |||
| 183 | LiteSeg-Darknet19 | 70.8% | cityscapes-dataset.com | Official leaderboard | |||
| 184 | SDBNet | 70.8% | cityscapes-dataset.com | Official leaderboard | |||
| 185 | MobileNetV2Plus | 70.7% | cityscapes-dataset.com | Official leaderboard | |||
| 186 | RFMobileNetV2Plus | 70.7% | cityscapes-dataset.com | Official leaderboard | |||
| 187 | AHiSS_ROB | 70.6% | cityscapes-dataset.com | Official leaderboard | |||
| 188 | SFRSeg | 70.6% | cityscapes-dataset.com | Official leaderboard | |||
| 189 | DeepLabv2-CRF | 70.4% | cityscapes-dataset.com | Official leaderboard | |||
| 190 | GUNet | 70.4% | cityscapes-dataset.com | Official leaderboard | |||
| 191 | ShuffleNet v2 + DPC | 70.3% | cityscapes-dataset.com | Official leaderboard | |||
| 192 | F2MF-short | 70.2% | cityscapes-dataset.com | Official leaderboard | |||
| 193 | AttaNet_light | 70.1% | cityscapes-dataset.com | Official leaderboard | |||
| 194 | CFPNet | 70.1% | cityscapes-dataset.com | Official leaderboard | |||
| 195 | SAITv2 | 70% | cityscapes-dataset.com | Official leaderboard | |||
| 196 | ERINet | 69.8% | cityscapes-dataset.com | Official leaderboard | |||
| 197 | EVANet | 69.8% | cityscapes-dataset.com | Official leaderboard | |||
| 198 | GridNet | 69.8% | Submission 653 | cityscapes-dataset.com | Official leaderboard | ||
| 199 | ERFNet (pretrained) | 69.7% | cityscapes-dataset.com | Official leaderboard | |||
| 200 | FSFFNet | 69.4% | cityscapes-dataset.com | Official leaderboard | |||
| 201 | SPSSN | 69.4% | cityscapes-dataset.com | Official leaderboard | |||
| 202 | DSNet | 69.3% | Submission 2621 | cityscapes-dataset.com | Official leaderboard | ||
| 203 | FSFNet | 69.1% | cityscapes-dataset.com | Official leaderboard | |||
| 204 | PL-Seg | 69.1% | cityscapes-dataset.com | Official leaderboard | |||
| 205 | Spatial Sampling Net | 68.9% | cityscapes-dataset.com | Official leaderboard | |||
| 206 | FarSee-Net | 68.4% | cityscapes-dataset.com | Official leaderboard | |||
| 207 | M2FANet | 68.3% | cityscapes-dataset.com | Official leaderboard | |||
| 208 | RPNet | 68.3% | cityscapes-dataset.com | Official leaderboard | |||
| 209 | SCMNet coarse | 68.3% | cityscapes-dataset.com | Official leaderboard | |||
| 210 | firenet | 68.2% | cityscapes-dataset.com | Official leaderboard | |||
| 211 | ERFNet (from scratch) | 68% | cityscapes-dataset.com | Official leaderboard | |||
| 212 | Fast-SCNN | 68% | cityscapes-dataset.com | Official leaderboard | |||
| 213 | MiniNet-v2-pretrained | 68% | cityscapes-dataset.com | Official leaderboard | |||
| 214 | SCMNet | 67.9% | cityscapes-dataset.com | Official leaderboard | |||
| 215 | LiteSeg-Mobilenet | 67.8% | cityscapes-dataset.com | Official leaderboard | |||
| 216 | MiniNet-v2-coarse | 67.8% | cityscapes-dataset.com | Official leaderboard | |||
| 217 | ERSNet-coarse | 67.6% | cityscapes-dataset.com | Official leaderboard | |||
| 218 | ERINet-v2 | 67.4% | cityscapes-dataset.com | Official leaderboard | |||
| 219 | NVSegNet | 67.4% | cityscapes-dataset.com | Official leaderboard | |||
| 220 | RGB-D FCN | 67.4% | cityscapes-dataset.com | Official leaderboard | |||
| 221 | EDANet | 67.3% | cityscapes-dataset.com | Official leaderboard | |||
| 222 | pruned & dilated inception-resnet-v2 (PD-IR2) | 67.3% | cityscapes-dataset.com | Official leaderboard | |||
| 223 | Dilation10 | 67.1% | cityscapes-dataset.com | Official leaderboard | |||
| 224 | seamseg_rvcsubset | 67% | cityscapes-dataset.com | Official leaderboard | |||
| 225 | DPN | 66.8% | Submission 82 | cityscapes-dataset.com | Official leaderboard | ||
| 226 | VENUS_ROB | 66.4% | cityscapes-dataset.com | Official leaderboard | |||
| 227 | Scale invariant CNN + CRF | 66.3% | cityscapes-dataset.com | Official leaderboard | |||
| 228 | ESPNetv2 | 66.2% | cityscapes-dataset.com | Official leaderboard | |||
| 229 | ContextNet | 66.1% | cityscapes-dataset.com | Official leaderboard | |||
| 230 | Le_Selfdriving_VGG | 65.9% | cityscapes-dataset.com | Official leaderboard | |||
| 231 | DeepLabv3+ decoder | 65.8% | TUE-5LSM0-g23 | cityscapes-dataset.com | Official leaderboard | ||
| 232 | FreeNet | 65.8% | cityscapes-dataset.com | Official leaderboard | |||
| 233 | Hybrid Model | 65.8% | cityscapes-dataset.com | Official leaderboard | |||
| 234 | FCN 8s | 65.3% | cityscapes-dataset.com | Official leaderboard | |||
| 235 | LiteSeg-Shufflenet | 65.2% | cityscapes-dataset.com | Official leaderboard | |||
| 236 | C3Net | 64.8% | cityscapes-dataset.com | Official leaderboard | |||
| 237 | CGNet | 64.8% | cityscapes-dataset.com | Official leaderboard | |||
| 238 | DeepLab LargeFOV StrongWeak | 64.8% | cityscapes-dataset.com | Official leaderboard | |||
| 239 | ndnet | 64.8% | cityscapes-dataset.com | Official leaderboard | |||
| 240 | ADSCNet | 64.5% | cityscapes-dataset.com | Official leaderboard | |||
| 241 | DeepLab+DynamicCRF | 64.5% | cityscapes-dataset.com | Official leaderboard | |||
| 242 | RMNet | 64.5% | cityscapes-dataset.com | Official leaderboard | |||
| 243 | Pixel-level Encoding for Instance Segmentation | 64.3% | cityscapes-dataset.com | Official leaderboard | |||
| 244 | FANet | 64.1% | cityscapes-dataset.com | Official leaderboard | |||
| 245 | ThunderNet | 64% | cityscapes-dataset.com | Official leaderboard | |||
| 246 | Mobilenetv3-small-backbone real-time segmentation | 63.9% | cityscapes-dataset.com | Official leaderboard | |||
| 247 | AdapNetv2_ROB | 63.8% | cityscapes-dataset.com | Official leaderboard | |||
| 248 | CLRCNet | 63.3% | cityscapes-dataset.com | Official leaderboard | |||
| 249 | EffPS_b1bs4_RVC | 63.2% | cityscapes-dataset.com | Official leaderboard | |||
| 250 | DeepLab LargeFOV Strong | 63.1% | cityscapes-dataset.com | Official leaderboard | |||
| 251 | ENet with the Lovász-Softmax loss | 63.1% | cityscapes-dataset.com | Official leaderboard | |||
| 252 | BiSeNetV2-CityLabeler | 63% | cityscapes-dataset.com | Official leaderboard | |||
| 253 | GoogLeNet FCN | 63% | cityscapes-dataset.com | Official leaderboard | |||
| 254 | Fast-SCNN (Half-resolution) | 62.8% | cityscapes-dataset.com | Official leaderboard | |||
| 255 | VlocNet++_ROB | 62.7% | cityscapes-dataset.com | Official leaderboard | |||
| 256 | CRFasRNN | 62.5% | cityscapes-dataset.com | Official leaderboard | |||
| 257 | SkipNet-MobileNet | 61.5% | cityscapes-dataset.com | Official leaderboard | |||
| 258 | LBPSS | 61% | cityscapes-dataset.com | Official leaderboard | |||
| 259 | ESPNet | 60.3% | cityscapes-dataset.com | Official leaderboard | |||
| 260 | SQ | 59.8% | cityscapes-dataset.com | Official leaderboard | |||
| 261 | GoogLeNetV1_ROB | 59.6% | cityscapes-dataset.com | Official leaderboard | |||
| 262 | MultiBoost | 59.3% | cityscapes-dataset.com | Official leaderboard | |||
| 263 | F2MF-mid | 59.1% | cityscapes-dataset.com | Official leaderboard | |||
| 264 | HiSS_ROB | 58.9% | cityscapes-dataset.com | Official leaderboard | |||
| 265 | ENet | 58.3% | cityscapes-dataset.com | Official leaderboard | |||
| 266 | ShuffleSeg | 58.3% | cityscapes-dataset.com | Official leaderboard | |||
| 267 | FCN8s-QunjieYu | 57.4% | cityscapes-dataset.com | Official leaderboard | |||
| 268 | Segnet basic | 57% | cityscapes-dataset.com | Official leaderboard | |||
| 269 | APMoE_seg_ROB | 56.5% | cityscapes-dataset.com | Official leaderboard | |||
| 270 | BiSeNet-V10 | 56.2% | cityscapes-dataset.com | Official leaderboard | |||
| 271 | Segnet extended | 56.1% | cityscapes-dataset.com | Official leaderboard | |||
| 272 | BatMAN_ROB | 55.4% | cityscapes-dataset.com | Official leaderboard | |||
| 273 | Fast-SCNN (Quarter-resolution) | 51.9% | cityscapes-dataset.com | Official leaderboard | |||
| 274 | Deform_ResNet_Balanced | 51.3% | cityscapes-dataset.com | Official leaderboard | |||
| 275 | UFONet (half-resolution) | 48% | cityscapes-dataset.com | Official leaderboard | |||
| 276 | MiniNet | 40.7% | cityscapes-dataset.com | Official leaderboard | |||
| 277 | FCN101_ROB | 30.4% | cityscapes-dataset.com | Official leaderboard | |||
| 278 | RFLR | 30% | cityscapes-dataset.com | Official leaderboard |
No lab model card reports it yet
We read the headline results of 540 launch posts from 57 labs, and none of them list this benchmark.
More in Vision
- MMMU In use 11.5K college-level multimodal questions across 30 subjects. Multiple-choice and open accuracy.
Qwen 3.6 Plus 86%
- MMMU-Pro In use Harder MMMU: college-level multimodal questions with more options and vision-only inputs to remove text shortcuts. GPT-5.6 Sol 84.6%
- CharXiv In use Chart understanding on real arXiv figures. Reasoning questions require synthesizing information from complex charts. Claude Fable 5 93.5%
- MathVista In use Math reasoning in visual contexts (charts, geometry, puzzles) from 31 datasets. Answer accuracy.
Seed2.1 Pro 90.7%
- MATH-Vision (MathVision) In use 3,040 competition math problems with visual context. Answer accuracy.
Qwen 3.8 Max 97.7%
- MMStar In use Models answer 1,500 human-curated image questions that cannot be solved without the image, across 6 core capabilities. Scored by accuracy.
Qwen 3.8 Max 85.9%
Data audit
All 8 checks pass. Every benchmark
Catalog record 2 pass
- Pass Task access All tasks are public.
- Pass Confidence Checked against a primary source.
Scores 2 pass 3 notes
- Pass Independent runs Scores from the official leaderboard (278 entries).
- Pass Models scored 278 entries on its leaderboard, none a model we track.
- Note Newest score No scores yet.
- Note Frontier models No models scored yet.
- Note Headroom No scores yet.
Leaderboard import 3 pass
- Pass Importer We read the board on cityscapes-dataset.com: 278 entries at the last read, none a model we track.
- Pass Last read November 20th, 2026.
- Pass Read worked The last read worked.
Consistency 1 pass 1 note
- Note Lab claims No model has both a lab number and an independent run.
- Pass Known issues No published audit found problems.