TasksSotADatasetsPapersMethodsSubmitAbout
Papers With Code 2

A community resource for machine learning research: papers, code, benchmarks, and state-of-the-art results.

Explore

Notable BenchmarksAll SotADatasetsPapersMethods

Community

Submit ResultsAbout

Data sourced from the PWC Archive (CC-BY-SA 4.0). Built by the community, for the community.

SotA/Computer Vision/Instance Segmentation/Cityscapes val

Instance Segmentation on Cityscapes val

Metric: mask AP (higher is better)

LeaderboardDataset
Loading chart...

Results

Submit a result
#Model↕mask AP▼Extra DataPaperDate↕Code
1ViT-P (OneFormer, ConvNeXt-L, single-scale, 512x1024, Mapillary Vistas-pretrained)49YesThe Missing Point in Vision Transformers for Uni...2025-05-26Code
2OneFormer (ConvNeXt-L, single-scale, Mapillary-Pretrained)48.7YesOneFormer: One Transformer to Rule Universal Ima...2022-11-10Code
3OpenSeeD( SwinL, single-scale)48.5YesA Simple Framework for Open-Vocabulary Segmentat...2023-03-14Code
4AFF-Base (single-scale, point-based Mask2Former)46.2NoAutoFocusFormer: Image Segmentation off the Grid2023-04-24Code
5OneFormer (DiNAT-L, single-scale)45.6NoOneFormer: One Transformer to Rule Universal Ima...2022-11-10Code
6OneFormer (Swin-L, single-scale)45.6NoOneFormer: One Transformer to Rule Universal Ima...2022-11-10Code
7DiNAT-L (single-scale, Mask2Former)45.1NoDilated Neighborhood Attention Transformer2022-09-29Code
8AFF-Small (single-scale, point-based Mask2Former)44NoAutoFocusFormer: Image Segmentation off the Grid2023-04-24Code
9Mask2Former (Swin-L, single-scale)43.7NoMasked-attention Mask Transformer for Universal ...2021-12-02Code
10Mask2Former (Swin-B)42NoMasked-attention Mask Transformer for Universal ...2021-12-02Code
11Mask2Former (Swin-S)41.8NoMasked-attention Mask Transformer for Universal ...2021-12-02Code
12PolySnake40.2NoRecurrent Generic Contour-based Instance Segment...2023-01-21Code
13Mask2Former (Swin-T)39.7NoMasked-attention Mask Transformer for Universal ...2021-12-02Code
14Mask2Former (ResNet-101)38.5NoMasked-attention Mask Transformer for Universal ...2021-12-02Code
15Mask2Former (ResNet-50)37.4NoMasked-attention Mask Transformer for Universal ...2021-12-02Code
16GAIS-Net37.1YesGeometry-Aware Instance Segmentation with Dispar...2020-06-14Code
17PointRend35.8NoPointRend: Image Segmentation as Rendering2019-12-17Code