Base-to-new Generalization on ModelNet40 (test)
95.3Base AccuracyULIP-2 + RC
Evaluation results
| Method | Links | |||
|---|---|---|---|---|
| ULIP-2 + RCprompt learning=true, regulation constraint framework=true2024.10 | 95.3 | 64.83 | 77.17 | |
| ULIP + RCprompt learning=true, regulation constraint framework=true2024.10 | 95.03 | 55.27 | 69.89 | |
| P-CLIP2prompt learning=true2024.10 | 93.98 | 45.21 | 61.05 | |
| P-CLIPprompt learning=true2024.10 | 93.23 | 20.22 | 33.23 | |
| ULIPprompt learning=true2024.10 | 92.8 | 50.07 | 65.05 | |
| ULIP-2prompt learning=true2024.10 | 91.77 | 56.47 | 69.92 |