SOTAVerified

Open Vocabulary Panoptic Segmentation

Papers

Showing 1117 of 17 papers

TitleStatusHype
Panoptic Vision-Language Feature FieldsCode1
Convolutions Die Hard: Open-Vocabulary Segmentation with Single Frozen Convolutional CLIPCode2
FreeSeg: Unified, Universal and Open-Vocabulary Image Segmentation0
Open-vocabulary Panoptic Segmentation with Embedding Modulation0
Open-Vocabulary Panoptic Segmentation with Text-to-Image Diffusion ModelsCode2
Open-Vocabulary Universal Image Segmentation with MaskCLIPCode1
Extract Free Dense Labels from CLIPCode1
Show:102550
← PrevPage 2 of 2Next →

Benchmark Results

#ModelMetricClaimedVerifiedStatus
1UMG-CLIP-E/14PQ31.6Unverified
2PosSAMPQ29.2Unverified
3UMG-CLIP-L/14PQ29.1Unverified
4MAFT+PQ27.1Unverified
5FC-CLIPPQ26.8Unverified
6CLIPSelfPQ23.7Unverified
7ODISE(Caption)PQ23.4Unverified
8ODISE (Label)PQ22.6Unverified
9FreeSegPQ16.3Unverified
10MaskCLIPPQ15.1Unverified