Tasks SotA Datasets Papers Methods Submit About

Papers With Code 2

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

Explore

Notable Benchmarks All SotA Datasets Papers Methods

Community

Submit Results About

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

Models/OneFormer (DiNAT-L, single-scale)

OneFormer (DiNAT-L, single-scale)

Reported on 41 benchmarks across 4 tasks · 1 paper · 7 SOTA

Note: results are matched by exact model name. Different papers may use the same name for different model variants.

Computer Vision15 results

Panoptic SegmentationonMapillary val
PQ· 2022-11-10
46.7
SOTA
OneFormer: One Transformer to Rule Universal Image Segmentation arXiv:2211.06220
Panoptic SegmentationonMapillary val
PQst· 2022-11-10
54.9
SOTA
OneFormer: One Transformer to Rule Universal Image Segmentation arXiv:2211.06220
Panoptic SegmentationonMapillary val
mIoU· 2022-11-10
61.7
SOTA
OneFormer: One Transformer to Rule Universal Image Segmentation arXiv:2211.06220
Instance SegmentationonCityscapes val
mask AP· 2022-11-10
45.6
best: 49 (ViT-P (OneFormer, ConvNeXt-L, single-scale, 512x1024, Mapillary Vistas-pretrained))
OneFormer: One Transformer to Rule Universal Image Segmentation arXiv:2211.06220
Instance SegmentationonCOCO val (panoptic labels)
AP· 2022-11-10
49.2
best: 52 (OneFormer (InternImage-H, emb_dim=1024, single-scale))
OneFormer: One Transformer to Rule Universal Image Segmentation arXiv:2211.06220
Instance SegmentationonADE20K val
AP· 2022-11-10
36
best: 44.2 (OneFormer (InternImage-H, emb_dim=1024, single-scale, 896x896, COCO-Pretrained))
OneFormer: One Transformer to Rule Universal Image Segmentation arXiv:2211.06220
Panoptic SegmentationonCityscapes val
AP· 2022-11-10
45.6
best: 50.6 (ViT-P (OneFormer, InternImage-H))
OneFormer: One Transformer to Rule Universal Image Segmentation arXiv:2211.06220
Panoptic SegmentationonCityscapes val
PQ· 2022-11-10
67.6
best: 70.8 (ViT-P (OneFormer, InternImage-H))
OneFormer: One Transformer to Rule Universal Image Segmentation arXiv:2211.06220
Panoptic SegmentationonCityscapes val
mIoU· 2022-11-10
83.1
best: 90.3 (EfficientPS (Cityscapes-fine))
OneFormer: One Transformer to Rule Universal Image Segmentation arXiv:2211.06220
Panoptic SegmentationonMapillary val
PQth· 2022-11-10
40.5
best: 40.8 (Panoptic FCN* (Swin-L, single-scale))
OneFormer: One Transformer to Rule Universal Image Segmentation arXiv:2211.06220
Panoptic SegmentationonCOCO minival
AP· 2022-11-10
49.2
best: 53.2 (OpenSeeD (SwinL, single-scale))
OneFormer: One Transformer to Rule Universal Image Segmentation arXiv:2211.06220
Panoptic SegmentationonCOCO minival
PQ· 2022-11-10
58
best: 61.2 (HyperSeg (Swin-B))
OneFormer: One Transformer to Rule Universal Image Segmentation arXiv:2211.06220
Panoptic SegmentationonCOCO minival
PQst· 2022-11-10
48.4
best: 49.2 (OneFormer (InternImage-H,single-scale))
OneFormer: One Transformer to Rule Universal Image Segmentation arXiv:2211.06220
Panoptic SegmentationonCOCO minival
PQth· 2022-11-10
64.3
best: 67.1 (OneFormer (InternImage-H,single-scale))
OneFormer: One Transformer to Rule Universal Image Segmentation arXiv:2211.06220
Panoptic SegmentationonCOCO minival
mIoU· 2022-11-10
68.1
best: 69.7 (UMG-CLIP-E/14)
OneFormer: One Transformer to Rule Universal Image Segmentation arXiv:2211.06220

Medical13 results

Semantic SegmentationonMapillary val
PQ· 2022-11-10
46.7
SOTA
OneFormer: One Transformer to Rule Universal Image Segmentation arXiv:2211.06220
Semantic SegmentationonMapillary val
PQst· 2022-11-10
54.9
SOTA
OneFormer: One Transformer to Rule Universal Image Segmentation arXiv:2211.06220
Semantic SegmentationonCOCO (Common Objects in Context)
mIoU· 2022-11-10
68.1
best: 77.2 (HyperSeg)
OneFormer: One Transformer to Rule Universal Image Segmentation arXiv:2211.06220
Semantic SegmentationonCityscapes val
AP· 2022-11-10
45.6
best: 50.6 (ViT-P (OneFormer, InternImage-H))
OneFormer: One Transformer to Rule Universal Image Segmentation arXiv:2211.06220
Semantic SegmentationonCityscapes val
PQ· 2022-11-10
67.6
best: 70.8 (ViT-P (OneFormer, InternImage-H))
OneFormer: One Transformer to Rule Universal Image Segmentation arXiv:2211.06220
Semantic SegmentationonCityscapes val
mIoU· 2022-11-10
83.1
best: 90.3 (EfficientPS (Cityscapes-fine))
OneFormer: One Transformer to Rule Universal Image Segmentation arXiv:2211.06220
Semantic SegmentationonMapillary val
PQth· 2022-11-10
40.5
best: 40.8 (Panoptic FCN* (Swin-L, single-scale))
OneFormer: One Transformer to Rule Universal Image Segmentation arXiv:2211.06220
Semantic SegmentationonMapillary val
mIoU· 2022-11-10
61.7
best: 76 (AO-SegNet)
OneFormer: One Transformer to Rule Universal Image Segmentation arXiv:2211.06220
Semantic SegmentationonCOCO minival
AP· 2022-11-10
49.2
best: 53.2 (OpenSeeD (SwinL, single-scale))
OneFormer: One Transformer to Rule Universal Image Segmentation arXiv:2211.06220
Semantic SegmentationonCOCO minival
PQ· 2022-11-10
58
best: 61.2 (HyperSeg (Swin-B))
OneFormer: One Transformer to Rule Universal Image Segmentation arXiv:2211.06220
Semantic SegmentationonCOCO minival
PQst· 2022-11-10
48.4
best: 49.2 (OneFormer (InternImage-H,single-scale))
OneFormer: One Transformer to Rule Universal Image Segmentation arXiv:2211.06220
Semantic SegmentationonCOCO minival
PQth· 2022-11-10
64.3
best: 67.1 (OneFormer (InternImage-H,single-scale))
OneFormer: One Transformer to Rule Universal Image Segmentation arXiv:2211.06220
Semantic SegmentationonCOCO minival
mIoU· 2022-11-10
68.1
best: 69.7 (UMG-CLIP-E/14)
OneFormer: One Transformer to Rule Universal Image Segmentation arXiv:2211.06220

Audio13 results

10-shot image generationonMapillary val
PQ· 2022-11-10
46.7
SOTA
OneFormer: One Transformer to Rule Universal Image Segmentation arXiv:2211.06220
10-shot image generationonMapillary val
PQst· 2022-11-10
54.9
SOTA
OneFormer: One Transformer to Rule Universal Image Segmentation arXiv:2211.06220
10-shot image generationonCOCO (Common Objects in Context)
mIoU· 2022-11-10
68.1
best: 77.2 (HyperSeg)
OneFormer: One Transformer to Rule Universal Image Segmentation arXiv:2211.06220
10-shot image generationonCityscapes val
AP· 2022-11-10
45.6
best: 50.6 (ViT-P (OneFormer, InternImage-H))
OneFormer: One Transformer to Rule Universal Image Segmentation arXiv:2211.06220
10-shot image generationonCityscapes val
PQ· 2022-11-10
67.6
best: 70.8 (ViT-P (OneFormer, InternImage-H))
OneFormer: One Transformer to Rule Universal Image Segmentation arXiv:2211.06220
10-shot image generationonCityscapes val
mIoU· 2022-11-10
83.1
best: 90.3 (EfficientPS (Cityscapes-fine))
OneFormer: One Transformer to Rule Universal Image Segmentation arXiv:2211.06220
10-shot image generationonMapillary val
PQth· 2022-11-10
40.5
best: 40.8 (Panoptic FCN* (Swin-L, single-scale))
OneFormer: One Transformer to Rule Universal Image Segmentation arXiv:2211.06220
10-shot image generationonMapillary val
mIoU· 2022-11-10
61.7
best: 76 (AO-SegNet)
OneFormer: One Transformer to Rule Universal Image Segmentation arXiv:2211.06220
10-shot image generationonCOCO minival
AP· 2022-11-10
49.2
best: 53.2 (OpenSeeD (SwinL, single-scale))
OneFormer: One Transformer to Rule Universal Image Segmentation arXiv:2211.06220
10-shot image generationonCOCO minival
PQ· 2022-11-10
58
best: 61.2 (HyperSeg (Swin-B))
OneFormer: One Transformer to Rule Universal Image Segmentation arXiv:2211.06220
10-shot image generationonCOCO minival
PQst· 2022-11-10
48.4
best: 49.2 (OneFormer (InternImage-H,single-scale))
OneFormer: One Transformer to Rule Universal Image Segmentation arXiv:2211.06220
10-shot image generationonCOCO minival
PQth· 2022-11-10
64.3
best: 67.1 (OneFormer (InternImage-H,single-scale))
OneFormer: One Transformer to Rule Universal Image Segmentation arXiv:2211.06220
10-shot image generationonCOCO minival
mIoU· 2022-11-10
68.1
best: 69.7 (UMG-CLIP-E/14)
OneFormer: One Transformer to Rule Universal Image Segmentation arXiv:2211.06220