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/TokenCut

TokenCut

Reported on 21 benchmarks across 6 tasks · 2 papers · 12 SOTA

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

Computer Vision21 results

Saliency DetectiononECSSD
Accuracy· 2022-02-23
93.4
best: 95.6 (MOVE)
SOTA
Self-Supervised Transformers for Unsupervised Object Discovery using Normalized Cut arXiv:2202.11539
Saliency DetectiononECSSD
IoU· 2022-02-23
77.2
best: 83.6 (MOVE)
SOTA
Self-Supervised Transformers for Unsupervised Object Discovery using Normalized Cut arXiv:2202.11539
Saliency DetectiononECSSD
maximal F-measure· 2022-02-23
87.4
best: 95.6 (SelfMask)
SOTA
Self-Supervised Transformers for Unsupervised Object Discovery using Normalized Cut arXiv:2202.11539
Saliency DetectiononDUT-OMRON
Accuracy· 2022-02-23
89.7
best: 93.7 (MOVE)
SOTA
Self-Supervised Transformers for Unsupervised Object Discovery using Normalized Cut arXiv:2202.11539
Saliency DetectiononDUT-OMRON
IoU· 2022-02-23
61.8
best: 66.6 (MOVE)
SOTA
Self-Supervised Transformers for Unsupervised Object Discovery using Normalized Cut arXiv:2202.11539
Saliency DetectiononDUT-OMRON
maximal F-measure· 2022-02-23
69.7
best: 85.2 (SelfMask)
SOTA
Self-Supervised Transformers for Unsupervised Object Discovery using Normalized Cut arXiv:2202.11539
Saliency DetectiononDUTS
Accuracy· 2022-02-23
91.4
best: 95.4 (MOVE)
SOTA
Self-Supervised Transformers for Unsupervised Object Discovery using Normalized Cut arXiv:2202.11539
Saliency DetectiononDUTS
IoU· 2022-02-23
62.4
best: 72.8 (MOVE)
SOTA
Self-Supervised Transformers for Unsupervised Object Discovery using Normalized Cut arXiv:2202.11539
Saliency DetectiononDUTS
maximal F-measure· 2022-02-23
75.5
best: 88.2 (SelfMask)
SOTA
Self-Supervised Transformers for Unsupervised Object Discovery using Normalized Cut arXiv:2202.11539
Object LocalizationonImageNet
GT-known localization accuracy· 2022-02-23
65.4
best: 75 (Stable diffusion)
SOTA
Self-Supervised Transformers for Unsupervised Object Discovery using Normalized Cut arXiv:2202.11539
Object LocalizationonImageNet
Top-1 Localization Accuracy· 2022-02-23
52.3
best: 65.2 (Stable diffusion)
SOTA
Self-Supervised Transformers for Unsupervised Object Discovery using Normalized Cut arXiv:2202.11539
Object LocalizationonCUB
Top-1 Localization Accuracy· 2022-02-23
72.9
SOTA
Self-Supervised Transformers for Unsupervised Object Discovery using Normalized Cut arXiv:2202.11539
Instance SegmentationonSegTrack-v2
mIoU· 2022-09-01
59.6
best: 79.6 (RCF (with post-processing))
TokenCut: Segmenting Objects in Images and Videos with Self-supervised Transformer and Normalized Cut arXiv:2209.00383
Instance SegmentationonFBMS-59
mIoU· 2022-09-01
60.2
best: 72.4 (RCF (with post-processing))
TokenCut: Segmenting Objects in Images and Videos with Self-supervised Transformer and Normalized Cut arXiv:2209.00383
Unsupervised Object SegmentationonSegTrack-v2
mIoU· 2022-09-01
59.6
best: 79.6 (RCF (with post-processing))
TokenCut: Segmenting Objects in Images and Videos with Self-supervised Transformer and Normalized Cut arXiv:2209.00383
Unsupervised Object SegmentationonFBMS-59
mIoU· 2022-09-01
60.2
best: 72.4 (RCF (with post-processing))
TokenCut: Segmenting Objects in Images and Videos with Self-supervised Transformer and Normalized Cut arXiv:2209.00383
Unsupervised Instance SegmentationonCOCO val2017
AP· 2022-09-01
2.4
best: 10.2 (CutS3D (DiffNCuts))
TokenCut: Segmenting Objects in Images and Videos with Self-supervised Transformer and Normalized Cut arXiv:2209.00383
Unsupervised Instance SegmentationonCOCO val2017
AP50· 2022-09-01
4.8
best: 20.1 (CutS3D (DiffNCuts))
TokenCut: Segmenting Objects in Images and Videos with Self-supervised Transformer and Normalized Cut arXiv:2209.00383
Unsupervised Instance SegmentationonCOCO val2017
AP75· 2022-09-01
1.9
best: 9.7 (CutLER (Cascade+DINO))
TokenCut: Segmenting Objects in Images and Videos with Self-supervised Transformer and Normalized Cut arXiv:2209.00383
Object Localizationon CUB-200-2011
Top-1 Localization Accuracy· 2022-02-23
72.9
best: 87 (GenPromp)
Self-Supervised Transformers for Unsupervised Object Discovery using Normalized Cut arXiv:2202.11539
Single-object discoveryonCOCO_20k
CorLoc· 2022-02-23
58.8
best: 72.2 (IMST)
Self-Supervised Transformers for Unsupervised Object Discovery using Normalized Cut arXiv:2202.11539