CLIP Surgery (original CLIP without any fine-tuning)

Reported on 2 benchmarks across 1 task · 1 paper · 1 SOTA

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

Computer Vision2 results

Open Vocabulary Semantic SegmentationonCOCO-Stuff-171
mIoU· 2023-04-12
21.9
best: 23.7 (TTD (TCL))
SOTA
A Closer Look at the Explainability of Contrastive Language-Image Pre-training arXiv:2304.05653
Open Vocabulary Semantic SegmentationonPASCAL Context-59
mIoU· 2023-04-12
29.3
best: 64.6 (HyperSeg)
A Closer Look at the Explainability of Contrastive Language-Image Pre-training arXiv:2304.05653