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/ReCo

ReCo

Reported on 46 benchmarks across 6 tasks · 2 papers · 15 SOTA

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

Medical18 results

Image GenerationonCOCO-MIG
instance success rate· 2022-11-23
0.55
best: 0.66 (MIGC)
SOTA
ReCo: Region-Controlled Text-to-Image Generation arXiv:2211.15518
Image GenerationonCOCO-MIG
mIoU· 2022-11-23
0.49
best: 0.56 (MIGC)
SOTA
ReCo: Region-Controlled Text-to-Image Generation arXiv:2211.15518
Image GenerationonLayoutBench-COCO - Size
AP· 2022-11-23
24.1
best: 33.5 (IterInpaint)
SOTA
ReCo: Region-Controlled Text-to-Image Generation arXiv:2211.15518
Image GenerationonLayoutBench-COCO - Combination
AP· 2022-11-23
18.7
best: 44.1 (IterInpaint)
SOTA
ReCo: Region-Controlled Text-to-Image Generation arXiv:2211.15518
Image GenerationonLayoutBench-COCO - Number
AP· 2022-11-23
30.9
best: 31.4 (IterInpaint)
SOTA
ReCo: Region-Controlled Text-to-Image Generation arXiv:2211.15518
Image GenerationonLayoutBench-COCO - Position
AP· 2022-11-23
36.4
best: 39.1 (IterInpaint)
SOTA
ReCo: Region-Controlled Text-to-Image Generation arXiv:2211.15518
Semantic SegmentationonADE20K
Mean IoU (val)· 2022-06-14
11.2
best: 30.7 (CorrCLIP)
SOTA
ReCo: Retrieve and Co-segment for Zero-shot Transfer arXiv:2206.07045
Image GenerationonLayoutBench
AP· 2022-11-23
7.6
best: 36.5 (IterInpaint)
ReCo: Region-Controlled Text-to-Image Generation arXiv:2211.15518
Semantic SegmentationonCOCO-Stuff-171
mIoU· 2022-06-14
14.8
best: 34 (CorrCLIP)
ReCo: Retrieve and Co-segment for Zero-shot Transfer arXiv:2206.07045
Semantic SegmentationonCOCO-Object
mIoU· 2022-06-14
15.7
best: 49.4 (CorrCLIP)
ReCo: Retrieve and Co-segment for Zero-shot Transfer arXiv:2206.07045
Semantic SegmentationonCityscapes val
mIoU· 2022-06-14
19.3
best: 90.3 (EfficientPS (Cityscapes-fine))
ReCo: Retrieve and Co-segment for Zero-shot Transfer arXiv:2206.07045
Semantic SegmentationonCityscapes val
pixel accuracy· 2022-06-14
74.6
best: 83.7 (ReCo+)
ReCo: Retrieve and Co-segment for Zero-shot Transfer arXiv:2206.07045
Semantic SegmentationonPASCAL Context-59
mIoU· 2022-06-14
22.3
best: 50.8 (CorrCLIP)
ReCo: Retrieve and Co-segment for Zero-shot Transfer arXiv:2206.07045
Semantic SegmentationonPascalVOC-20
mIoU· 2022-06-14
57.7
best: 91.8 (CorrCLIP)
ReCo: Retrieve and Co-segment for Zero-shot Transfer arXiv:2206.07045
Semantic SegmentationonKITTI-STEP
mIoU· 2022-06-14
29.8
best: 31.9 (ReCo+)
ReCo: Retrieve and Co-segment for Zero-shot Transfer arXiv:2206.07045
Semantic SegmentationonKITTI-STEP
pixel accuracy· 2022-06-14
70.6
best: 75.3 (ReCo+)
ReCo: Retrieve and Co-segment for Zero-shot Transfer arXiv:2206.07045
Semantic SegmentationonCOCO-Stuff-27
mIoU· 2022-06-14
26.3
best: 43.6 (DiffSeg (512))
ReCo: Retrieve and Co-segment for Zero-shot Transfer arXiv:2206.07045
Semantic SegmentationonCOCO-Stuff-27
pixel accuracy· 2022-06-14
46.1
best: 54.1 (ReCo+)
ReCo: Retrieve and Co-segment for Zero-shot Transfer arXiv:2206.07045

Audio15 results

10-shot image generationonCOCO-MIG
instance success rate· 2022-11-23
0.55
best: 0.66 (MIGC)
SOTA
ReCo: Region-Controlled Text-to-Image Generation arXiv:2211.15518
10-shot image generationonCOCO-MIG
mIoU· 2022-11-23
0.49
best: 0.56 (MIGC)
SOTA
ReCo: Region-Controlled Text-to-Image Generation arXiv:2211.15518
1 Image, 2*2 StitchionCOCO-MIG
instance success rate· 2022-11-23
0.55
best: 0.66 (MIGC)
SOTA
ReCo: Region-Controlled Text-to-Image Generation arXiv:2211.15518
1 Image, 2*2 StitchionCOCO-MIG
mIoU· 2022-11-23
0.49
best: 0.56 (MIGC)
SOTA
ReCo: Region-Controlled Text-to-Image Generation arXiv:2211.15518
10-shot image generationonADE20K
Mean IoU (val)· 2022-06-14
11.2
best: 30.7 (CorrCLIP)
SOTA
ReCo: Retrieve and Co-segment for Zero-shot Transfer arXiv:2206.07045
10-shot image generationonCOCO-Stuff-171
mIoU· 2022-06-14
14.8
best: 34 (CorrCLIP)
ReCo: Retrieve and Co-segment for Zero-shot Transfer arXiv:2206.07045
10-shot image generationonCOCO-Object
mIoU· 2022-06-14
15.7
best: 49.4 (CorrCLIP)
ReCo: Retrieve and Co-segment for Zero-shot Transfer arXiv:2206.07045
10-shot image generationonCityscapes val
mIoU· 2022-06-14
19.3
best: 90.3 (EfficientPS (Cityscapes-fine))
ReCo: Retrieve and Co-segment for Zero-shot Transfer arXiv:2206.07045
10-shot image generationonCityscapes val
pixel accuracy· 2022-06-14
74.6
best: 83.7 (ReCo+)
ReCo: Retrieve and Co-segment for Zero-shot Transfer arXiv:2206.07045
10-shot image generationonPASCAL Context-59
mIoU· 2022-06-14
22.3
best: 50.8 (CorrCLIP)
ReCo: Retrieve and Co-segment for Zero-shot Transfer arXiv:2206.07045
10-shot image generationonPascalVOC-20
mIoU· 2022-06-14
57.7
best: 91.8 (CorrCLIP)
ReCo: Retrieve and Co-segment for Zero-shot Transfer arXiv:2206.07045
10-shot image generationonKITTI-STEP
mIoU· 2022-06-14
29.8
best: 31.9 (ReCo+)
ReCo: Retrieve and Co-segment for Zero-shot Transfer arXiv:2206.07045
10-shot image generationonKITTI-STEP
pixel accuracy· 2022-06-14
70.6
best: 75.3 (ReCo+)
ReCo: Retrieve and Co-segment for Zero-shot Transfer arXiv:2206.07045
10-shot image generationonCOCO-Stuff-27
mIoU· 2022-06-14
26.3
best: 43.6 (DiffSeg (512))
ReCo: Retrieve and Co-segment for Zero-shot Transfer arXiv:2206.07045
10-shot image generationonCOCO-Stuff-27
pixel accuracy· 2022-06-14
46.1
best: 54.1 (ReCo+)
ReCo: Retrieve and Co-segment for Zero-shot Transfer arXiv:2206.07045

Computer Vision11 results

Unsupervised Semantic SegmentationonADE20K
Mean IoU (val)· 2022-06-14
11.2
best: 30.7 (CorrCLIP)
SOTA
ReCo: Retrieve and Co-segment for Zero-shot Transfer arXiv:2206.07045
Unsupervised Semantic SegmentationonCOCO-Stuff-171
mIoU· 2022-06-14
14.8
best: 34 (CorrCLIP)
ReCo: Retrieve and Co-segment for Zero-shot Transfer arXiv:2206.07045
Unsupervised Semantic SegmentationonCOCO-Object
mIoU· 2022-06-14
15.7
best: 49.4 (CorrCLIP)
ReCo: Retrieve and Co-segment for Zero-shot Transfer arXiv:2206.07045
Unsupervised Semantic SegmentationonCityscapes val
mIoU· 2022-06-14
19.3
best: 51.1 (CorrCLIP)
ReCo: Retrieve and Co-segment for Zero-shot Transfer arXiv:2206.07045
Unsupervised Semantic SegmentationonCityscapes val
pixel accuracy· 2022-06-14
74.6
best: 83.7 (ReCo+)
ReCo: Retrieve and Co-segment for Zero-shot Transfer arXiv:2206.07045
Unsupervised Semantic SegmentationonPASCAL Context-59
mIoU· 2022-06-14
22.3
best: 50.8 (CorrCLIP)
ReCo: Retrieve and Co-segment for Zero-shot Transfer arXiv:2206.07045
Unsupervised Semantic SegmentationonPascalVOC-20
mIoU· 2022-06-14
57.7
best: 91.8 (CorrCLIP)
ReCo: Retrieve and Co-segment for Zero-shot Transfer arXiv:2206.07045
Unsupervised Semantic SegmentationonKITTI-STEP
mIoU· 2022-06-14
29.8
best: 31.9 (ReCo+)
ReCo: Retrieve and Co-segment for Zero-shot Transfer arXiv:2206.07045
Unsupervised Semantic SegmentationonKITTI-STEP
pixel accuracy· 2022-06-14
70.6
best: 75.3 (ReCo+)
ReCo: Retrieve and Co-segment for Zero-shot Transfer arXiv:2206.07045
Unsupervised Semantic SegmentationonCOCO-Stuff-27
mIoU· 2022-06-14
26.3
best: 32.6 (ReCo+)
ReCo: Retrieve and Co-segment for Zero-shot Transfer arXiv:2206.07045
Unsupervised Semantic SegmentationonCOCO-Stuff-27
pixel accuracy· 2022-06-14
46.1
best: 54.1 (ReCo+)
ReCo: Retrieve and Co-segment for Zero-shot Transfer arXiv:2206.07045

Natural Language Processing2 results

Text-to-Image GenerationonCOCO-MIG
instance success rate· 2022-11-23
0.55
best: 0.66 (MIGC)
SOTA
ReCo: Region-Controlled Text-to-Image Generation arXiv:2211.15518
Text-to-Image GenerationonCOCO-MIG
mIoU· 2022-11-23
0.49
best: 0.56 (MIGC)
SOTA
ReCo: Region-Controlled Text-to-Image Generation arXiv:2211.15518