| 1 | VPNeXt | 71.1 | No | VPNeXt -- Rethinking Dense Decoding for Plain Vi... | 2025-02-23 | - |
| 2 | PlainSeg (EVA-02-L) | 71 | No | Minimalist and High-Performance Semantic Segment... | 2023-10-19 | Code |
| 3 | InternImage-H | 70.3 | No | InternImage: Exploring Large-Scale Vision Founda... | 2022-11-10 | Code |
| 4 | RSSeg-ViT-L (BEiT pretrain) | 68.9 | No | Representation Separation for Semantic Segmentat... | 2022-12-28 | - |
| 5 | ViT-Adapter-L (Mask2Former, BEiT pretrain) | 68.2 | No | Vision Transformer Adapter for Dense Predictions | 2022-05-17 | Code |
| 6 | ViT-Adapter-L (UperNet, BEiT pretrain) | 67.5 | No | Vision Transformer Adapter for Dense Predictions | 2022-05-17 | Code |
| 7 | RSSeg-ViT-L | 67.5 | No | Representation Separation for Semantic Segmentat... | 2022-12-28 | - |
| 8 | SegViT (ours) | 65.3 | No | SegViT: Semantic Segmentation with Plain Vision ... | 2022-10-12 | Code |
| 9 | CAA + CAR (ConvNeXt-Large + JPU) | 64.1 | No | CAR: Class-aware Regularizations for Semantic Se... | 2022-03-14 | Code |
| 10 | SenFormer (Swin-L) | 64 | No | Efficient Self-Ensemble for Semantic Segmentation | 2021-11-26 | Code |
| 11 | Sequential Ensemble (Segformer + HRNet) | 62.1 | No | Sequential Ensembling for Semantic Segmentation | 2022-10-08 | - |
| 12 | CAA + Simple decoder (Efficientnet-B7) | 60.5 | No | Channelized Axial Attention for Semantic Segment... | 2021-01-19 | Code |
| 13 | DPT-Hybrid | 60.46 | No | Vision Transformers for Dense Prediction | 2021-03-24 | Code |
| 14 | CAA (Efficientnet-B7) | 60.1 | No | Channelized Axial Attention for Semantic Segment... | 2021-01-19 | Code |
| 15 | HRNetV2 + OCR + RMI (PaddleClas pretrained) | 59.6 | No | Segmentation Transformer: Object-Contextual Repr... | 2019-09-24 | Code |
| 16 | Seg-L-Mask/16 | 59 | No | Segmenter: Transformer for Semantic Segmentation | 2021-05-12 | Code |
| 17 | ResNeSt-269 | 58.9 | No | ResNeSt: Split-Attention Networks | 2020-04-19 | Code |
| 18 | DEPICT-SA (ViT-L multi-scale) | 58.6 | No | Rethinking Decoders for Transformer-based Semant... | 2024-11-05 | Code |
| 19 | ResNeSt-200 | 58.4 | No | ResNeSt: Split-Attention Networks | 2020-04-19 | Code |
| 20 | DEPICT-SA (ViT-L single-scale) | 57.9 | No | Rethinking Decoders for Transformer-based Semant... | 2024-11-05 | Code |
| 21 | CondNet(ResNest-101) | 57 | No | CondNet: Conditional Classifier for Scene Segmen... | 2021-09-21 | Code |
| 22 | SenFormer (ResNet-101) | 56.6 | No | Efficient Self-Ensemble for Semantic Segmentation | 2021-11-26 | Code |
| 23 | ResNeSt-101 | 56.5 | No | ResNeSt: Split-Attention Networks | 2020-04-19 | Code |
| 24 | OCR (HRNetV2-W48) | 56.2 | No | Segmentation Transformer: Object-Contextual Repr... | 2019-09-24 | Code |
| 25 | GPaCo (ResNet101) | 56.2 | No | Generalized Parametric Contrastive Learning | 2022-09-26 | Code |
| 26 | CondNet(ResNet-101) | 56 | No | CondNet: Conditional Classifier for Scene Segmen... | 2021-09-21 | Code |
| 27 | SETR-MLA (16, 80k, MS) | 55.83 | No | Rethinking Semantic Segmentation from a Sequence... | 2020-12-31 | Code |
| 28 | DCNAS | 55.6 | No | DCNAS: Densely Connected Neural Architecture Sea... | 2020-03-26 | - |
| 29 | DNL | 55.3 | No | Disentangled Non-Local Neural Networks | 2020-06-11 | Code |
| 30 | HamNet (ResNet-101) | 55.2 | No | Is Attention Better Than Matrix Decomposition? | 2021-09-09 | Code |
| 31 | CAA (ResNet-101) | 55 | No | Channelized Axial Attention for Semantic Segment... | 2021-01-19 | Code |
| 32 | OCR (ResNet-101) | 54.8 | No | Segmentation Transformer: Object-Contextual Repr... | 2019-09-24 | Code |
| 33 | SIW(Segformer-B5) | 54.2 | No | Scaling up Multi-domain Semantic Segmentation wi... | 2022-02-04 | - |
| 34 | CFNet (ResNet-101) | 54 | No | Deep High-Resolution Representation Learning for... | 2019-08-20 | Code |
| 35 | CFNet (ResNet-101) | 54 | No | Deep High-Resolution Representation Learning for... | 2019-08-20 | Code |
| 36 | HRNetV2 HRNetV2-W48 | 54 | No | Deep High-Resolution Representation Learning for... | 2019-08-20 | Code |
| 37 | CPN(ResNet-101) | 53.9 | No | Context Prior for Scene Segmentation | 2020-04-03 | Code |
| 38 | LaU-regression-loss (ResNet-101) | 53.9 | No | Location-aware Upsampling for Semantic Segmentat... | 2019-11-13 | Code |
| 39 | DGCNet (MS, ResNet-101) | 53.7 | No | Dual Graph Convolutional Network for Semantic Se... | 2019-09-13 | Code |
| 40 | BFP | 53.6 | No | Boundary-Aware Feature Propagation for Scene Seg... | 2019-08-31 | Code |
| 41 | SVCNet (ResNet-101) | 53.2 | No | Semantic Correlation Promoted Shape-Variant Cont... | 2019-09-05 | Code |
| 42 | Joint Pyramid Upsampling + EncNet | 53.1 | No | FastFCN: Rethinking Dilated Convolution in the B... | 2019-03-28 | Code |
| 43 | EMANet | 53.1 | No | Expectation-Maximization Attention Networks for ... | 2019-07-31 | Code |
| 44 | Asymmetric ALNN | 52.8 | No | Asymmetric Non-local Neural Networks for Semanti... | 2019-08-21 | Code |
| 45 | CASSOD | 52.76 | No | CASSOD-Net: Cascaded and Separable Structures of... | 2021-04-29 | - |
| 46 | DANet (ResNet-101) | 52.6 | No | Dual Attention Network for Scene Segmentation | 2018-09-09 | Code |
| 47 | ICM | 52.6 | No | - | - | Code |
| 48 | DUpsampling | 52.5 | No | Decoders Matter for Semantic Segmentation: Data-... | 2019-03-05 | - |
| 49 | EncNet (ResNet-101) | 51.7 | No | Context Encoding for Semantic Segmentation | 2018-03-23 | Code |
| 50 | CFNet (ResNet-50) | 51.5 | No | - | - | Code |
| 51 | ResNet-38 | 48.1 | No | Wider or Deeper: Revisiting the ResNet Model for... | 2016-11-30 | Code |
| 52 | PSPNet (ResNet-101) | 47.8 | No | Pyramid Scene Parsing Network | 2016-12-04 | Code |
| 53 | RefineNet | 47.3 | No | RefineNet: Multi-Path Refinement Networks for Hi... | 2016-11-20 | Code |
| 54 | DeepLabV2 | 45.7 | No | DeepLab: Semantic Image Segmentation with Deep C... | 2016-06-02 | Code |
| 55 | VeryDeep | 44.5 | No | Bridging Category-level and Instance-level Seman... | 2016-05-23 | - |
| 56 | Piecewise | 43.3 | No | Efficient piecewise training of deep structured ... | 2015-04-04 | - |
| 57 | Dilated-FCN2s | 42.6 | No | Efficient Yet Deep Convolutional Neural Networks... | 2017-07-26 | Code |
| 58 | HO CRF | 41.3 | No | Higher Order Conditional Random Fields in Deep N... | 2015-11-25 | Code |
| 59 | BoxSup | 40.5 | No | BoxSup: Exploiting Bounding Boxes to Supervise C... | 2015-03-05 | - |
| 60 | ParseNet | 40.4 | No | ParseNet: Looking Wider to See Better | 2015-06-15 | Code |
| 61 | CRF-RNN | 39.3 | No | Conditional Random Fields as Recurrent Neural Ne... | 2015-02-11 | Code |
| 62 | FCN-8s | 37.8 | No | Fully Convolutional Networks for Semantic Segmen... | 2014-11-14 | Code |
| 63 | CFM | 34.4 | No | Convolutional Feature Masking for Joint Object a... | 2014-12-03 | Code |
| 64 | RBE2E | 32.5 | No | Region-based semantic segmentation with end-to-e... | 2016-07-26 | Code |
| 65 | SegCLIP | 24.7 | No | SegCLIP: Patch Aggregation with Learnable Center... | 2022-11-27 | Code |