Open-Vocabulary Semantic Segmentation Efficiency Suite (test/val)
433.7Params (M)CAT-Seg
Evaluation Results
| Method | Links | |||
|---|---|---|---|---|
| CAT-SegVLM Backbone=CLIP ViT-B/16, GPU=A60002026.01 | 433.7 | 2,121.1 | 0.78 | |
| ESC-NetVLM Backbone=CLIP ViT-B/16, GPU=A60002026.01 | 451.3 | 2,203.5 | 0.76 | |
| DiSaVLM Backbone=CLIP ViT-B/16, GPU=A60002026.01 | 456.2 | 2,287.3 | 0.69 | |
| ZSsegVLM Backbone=CLIP ViT-B/16, GPU=A60002026.01 | 530.8 | 22,302.1 | 3.11 | |
| ZegFormerVLM Backbone=CLIP ViT-B/16, GPU=A60002026.01 | 531.2 | 19,425.6 | 3.1 | |
| OVSegVLM Backbone=CLIP ViT-B/16, GPU=A60002026.01 | 532.6 | 19,345.6 | 2.98 |