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/MaskCLIP

MaskCLIP

Reported on 32 benchmarks across 6 tasks · 2 papers · 29 SOTA

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

Computer Vision14 results

Open Vocabulary Semantic SegmentationonADE20K-847
mIoU· 2022-08-18
8.2
best: 17.3 (UMG-CLIP-E/14)
SOTA
Open-Vocabulary Universal Image Segmentation with MaskCLIP arXiv:2208.08984
Open Vocabulary Semantic SegmentationonADE20K-150
mIoU· 2022-08-18
23.7
best: 38.2 (Mask-Adapter)
SOTA
Open-Vocabulary Universal Image Segmentation with MaskCLIP arXiv:2208.08984
Open Vocabulary Panoptic SegmentationonADE20K
PQ· 2021-12-02
15.1
best: 31.6 (UMG-CLIP-E/14)
SOTA
Extract Free Dense Labels from CLIP arXiv:2112.01071
Zero Shot SegmentationonADE20K training-free zero-shot segmentation
mIoU· 2021-12-02
10.2
best: 17.7 (COSMOS ViT-B/16)
SOTA
Extract Free Dense Labels from CLIP arXiv:2112.01071
Unsupervised Semantic SegmentationonCOCO-Stuff-171
mIoU· 2021-12-02
16.4
best: 34 (CorrCLIP)
SOTA
Extract Free Dense Labels from CLIP arXiv:2112.01071
Unsupervised Semantic SegmentationonCOCO-Object
mIoU· 2021-12-02
20.6
best: 49.4 (CorrCLIP)
SOTA
Extract Free Dense Labels from CLIP arXiv:2112.01071
Unsupervised Semantic SegmentationonADE20K
Mean IoU (val)· 2021-12-02
9.8
best: 30.7 (CorrCLIP)
SOTA
Extract Free Dense Labels from CLIP arXiv:2112.01071
Unsupervised Semantic SegmentationonCityscapes val
mIoU· 2021-12-02
10
best: 51.1 (CorrCLIP)
SOTA
Extract Free Dense Labels from CLIP arXiv:2112.01071
Unsupervised Semantic SegmentationonCityscapes val
pixel accuracy· 2021-12-02
35.9
best: 83.7 (ReCo+)
SOTA
Extract Free Dense Labels from CLIP arXiv:2112.01071
Unsupervised Semantic SegmentationonPASCAL Context-59
mIoU· 2021-12-02
26.4
best: 50.8 (CorrCLIP)
SOTA
Extract Free Dense Labels from CLIP arXiv:2112.01071
Unsupervised Semantic SegmentationonPascalVOC-20
mIoU· 2021-12-02
74.9
best: 91.8 (CorrCLIP)
SOTA
Extract Free Dense Labels from CLIP arXiv:2112.01071
Unsupervised Semantic SegmentationonPASCAL VOC
mIoU· 2021-12-02
29.3
best: 76.7 (CorrCLIP)
SOTA
Extract Free Dense Labels from CLIP arXiv:2112.01071
Open Vocabulary Semantic SegmentationonPASCAL Context-459
mIoU· 2021-12-02
10
best: 25.8 (SILC)
SOTA
Extract Free Dense Labels from CLIP arXiv:2112.01071
Open Vocabulary Semantic SegmentationonPASCAL Context-59
mIoU· 2022-08-18
45.9
best: 64.6 (HyperSeg)
Open-Vocabulary Universal Image Segmentation with MaskCLIP arXiv:2208.08984

Medical9 results

Semantic SegmentationonCC3M-TagMask
mIoU· 2021-12-02
41
best: 65.5 (TTD (TCL))
SOTA
Extract Free Dense Labels from CLIP arXiv:2112.01071
Semantic SegmentationonCOCO-Stuff-171
mIoU· 2021-12-02
16.4
best: 34 (CorrCLIP)
SOTA
Extract Free Dense Labels from CLIP arXiv:2112.01071
Semantic SegmentationonCOCO-Object
mIoU· 2021-12-02
20.6
best: 49.4 (CorrCLIP)
SOTA
Extract Free Dense Labels from CLIP arXiv:2112.01071
Semantic SegmentationonADE20K
Mean IoU (val)· 2021-12-02
9.8
best: 30.7 (CorrCLIP)
SOTA
Extract Free Dense Labels from CLIP arXiv:2112.01071
Semantic SegmentationonCityscapes val
pixel accuracy· 2021-12-02
35.9
best: 83.7 (ReCo+)
SOTA
Extract Free Dense Labels from CLIP arXiv:2112.01071
Semantic SegmentationonPASCAL Context-59
mIoU· 2021-12-02
26.4
best: 50.8 (CorrCLIP)
SOTA
Extract Free Dense Labels from CLIP arXiv:2112.01071
Semantic SegmentationonPascalVOC-20
mIoU· 2021-12-02
74.9
best: 91.8 (CorrCLIP)
SOTA
Extract Free Dense Labels from CLIP arXiv:2112.01071
Semantic SegmentationonPASCAL VOC
mIoU· 2021-12-02
29.3
best: 76.7 (CorrCLIP)
SOTA
Extract Free Dense Labels from CLIP arXiv:2112.01071
Semantic SegmentationonCityscapes val
mIoU· 2021-12-02
10
best: 90.3 (EfficientPS (Cityscapes-fine))
Extract Free Dense Labels from CLIP arXiv:2112.01071

Audio9 results

10-shot image generationonCC3M-TagMask
mIoU· 2021-12-02
41
best: 65.5 (TTD (TCL))
SOTA
Extract Free Dense Labels from CLIP arXiv:2112.01071
10-shot image generationonCOCO-Stuff-171
mIoU· 2021-12-02
16.4
best: 34 (CorrCLIP)
SOTA
Extract Free Dense Labels from CLIP arXiv:2112.01071
10-shot image generationonCOCO-Object
mIoU· 2021-12-02
20.6
best: 49.4 (CorrCLIP)
SOTA
Extract Free Dense Labels from CLIP arXiv:2112.01071
10-shot image generationonADE20K
Mean IoU (val)· 2021-12-02
9.8
best: 30.7 (CorrCLIP)
SOTA
Extract Free Dense Labels from CLIP arXiv:2112.01071
10-shot image generationonCityscapes val
pixel accuracy· 2021-12-02
35.9
best: 83.7 (ReCo+)
SOTA
Extract Free Dense Labels from CLIP arXiv:2112.01071
10-shot image generationonPASCAL Context-59
mIoU· 2021-12-02
26.4
best: 50.8 (CorrCLIP)
SOTA
Extract Free Dense Labels from CLIP arXiv:2112.01071
10-shot image generationonPascalVOC-20
mIoU· 2021-12-02
74.9
best: 91.8 (CorrCLIP)
SOTA
Extract Free Dense Labels from CLIP arXiv:2112.01071
10-shot image generationonPASCAL VOC
mIoU· 2021-12-02
29.3
best: 76.7 (CorrCLIP)
SOTA
Extract Free Dense Labels from CLIP arXiv:2112.01071
10-shot image generationonCityscapes val
mIoU· 2021-12-02
10
best: 90.3 (EfficientPS (Cityscapes-fine))
Extract Free Dense Labels from CLIP arXiv:2112.01071