Text-level Semantic Segmentation on CC3M (subset)
65.5Caption IoUTCL + TTD
Evaluation Results
| Method | Links | |||
|---|---|---|---|---|
| TCL + TTDFine-tuning=TTD2024.03 | 65.5 | 0.163 | 0.182 | |
| TCL2024.03 | 60.4 | 0.199 | 0.198 | |
| MaskCLIP + TTDFine-tuning=TTD2024.03 | 50.2 | 0.256 | 0.242 | |
| MaskCLIP2024.03 | 41 | 0.179 | 0.411 |