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/UMG-CLIP-L/14

UMG-CLIP-L/14

Reported on 15 benchmarks across 5 tasks · 1 paper · 1 SOTA

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

Computer Vision9 results

Open Vocabulary Semantic SegmentationonPascalVOC-20
mIoU· 2024-01-12
97.9
SOTA
UMG-CLIP: A Unified Multi-Granularity Vision Generalist for Open-World Understanding arXiv:2401.06397
Open Vocabulary Panoptic SegmentationonADE20K
PQ· 2024-01-12
29.1
best: 31.6 (UMG-CLIP-E/14)
UMG-CLIP: A Unified Multi-Granularity Vision Generalist for Open-World Understanding arXiv:2401.06397
Open Vocabulary Semantic SegmentationonADE20K-847
mIoU· 2024-01-12
15.4
best: 17.3 (UMG-CLIP-E/14)
UMG-CLIP: A Unified Multi-Granularity Vision Generalist for Open-World Understanding arXiv:2401.06397
Open Vocabulary Semantic SegmentationonPASCAL Context-459
mIoU· 2024-01-12
23.2
best: 25.8 (SILC)
UMG-CLIP: A Unified Multi-Granularity Vision Generalist for Open-World Understanding arXiv:2401.06397
Open Vocabulary Semantic SegmentationonPASCAL Context-59
mIoU· 2024-01-12
61
best: 64.6 (HyperSeg)
UMG-CLIP: A Unified Multi-Granularity Vision Generalist for Open-World Understanding arXiv:2401.06397
Open Vocabulary Semantic SegmentationonADE20K-150
mIoU· 2024-01-12
36.1
best: 38.2 (Mask-Adapter)
UMG-CLIP: A Unified Multi-Granularity Vision Generalist for Open-World Understanding arXiv:2401.06397
Panoptic SegmentationonCOCO minival
AP· uses extra data· 2024-01-12
49.7
best: 53.2 (OpenSeeD (SwinL, single-scale))
UMG-CLIP: A Unified Multi-Granularity Vision Generalist for Open-World Understanding arXiv:2401.06397
Panoptic SegmentationonCOCO minival
PQ· uses extra data· 2024-01-12
58.9
best: 61.2 (HyperSeg (Swin-B))
UMG-CLIP: A Unified Multi-Granularity Vision Generalist for Open-World Understanding arXiv:2401.06397
Panoptic SegmentationonCOCO minival
mIoU· uses extra data· 2024-01-12
68.9
best: 69.7 (UMG-CLIP-E/14)
UMG-CLIP: A Unified Multi-Granularity Vision Generalist for Open-World Understanding arXiv:2401.06397

Medical3 results

Semantic SegmentationonCOCO minival
AP· uses extra data· 2024-01-12
49.7
best: 53.2 (OpenSeeD (SwinL, single-scale))
UMG-CLIP: A Unified Multi-Granularity Vision Generalist for Open-World Understanding arXiv:2401.06397
Semantic SegmentationonCOCO minival
PQ· uses extra data· 2024-01-12
58.9
best: 61.2 (HyperSeg (Swin-B))
UMG-CLIP: A Unified Multi-Granularity Vision Generalist for Open-World Understanding arXiv:2401.06397
Semantic SegmentationonCOCO minival
mIoU· uses extra data· 2024-01-12
68.9
best: 69.7 (UMG-CLIP-E/14)
UMG-CLIP: A Unified Multi-Granularity Vision Generalist for Open-World Understanding arXiv:2401.06397

Audio3 results

10-shot image generationonCOCO minival
AP· uses extra data· 2024-01-12
49.7
best: 53.2 (OpenSeeD (SwinL, single-scale))
UMG-CLIP: A Unified Multi-Granularity Vision Generalist for Open-World Understanding arXiv:2401.06397
10-shot image generationonCOCO minival
PQ· uses extra data· 2024-01-12
58.9
best: 61.2 (HyperSeg (Swin-B))
UMG-CLIP: A Unified Multi-Granularity Vision Generalist for Open-World Understanding arXiv:2401.06397
10-shot image generationonCOCO minival
mIoU· uses extra data· 2024-01-12
68.9
best: 69.7 (UMG-CLIP-E/14)
UMG-CLIP: A Unified Multi-Granularity Vision Generalist for Open-World Understanding arXiv:2401.06397