LaVG

Reported on 4 benchmarks across 1 task · 1 paper

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

Computer Vision4 results

Open Vocabulary Semantic SegmentationonCOCO-Stuff-171
mIoU· 2024-08-09
23.2
best: 23.7 (TTD (TCL))
In Defense of Lazy Visual Grounding for Open-Vocabulary Semantic Segmentation arXiv:2408.04961
Open Vocabulary Semantic SegmentationonPascalVOC-20
mIoU· 2024-08-09
82.5
best: 97.9 (UMG-CLIP-L/14)
In Defense of Lazy Visual Grounding for Open-Vocabulary Semantic Segmentation arXiv:2408.04961
Open Vocabulary Semantic SegmentationonPASCAL Context-59
mIoU· 2024-08-09
34.7
best: 64.6 (HyperSeg)
In Defense of Lazy Visual Grounding for Open-Vocabulary Semantic Segmentation arXiv:2408.04961
Open Vocabulary Semantic SegmentationonADE20K-150
mIoU· 2024-08-09
15.8
best: 38.2 (Mask-Adapter)
In Defense of Lazy Visual Grounding for Open-Vocabulary Semantic Segmentation arXiv:2408.04961