In-Image Machine Translation Evaluation on Vectra-Bench Overall 1.0 (in-domain)
0.738Pearson Correlation (r)Vectra
Evaluation Results
| Method | Links | ||
|---|---|---|---|
| VectraProtocol=No-Reference (Direct Assessment)2026.01 | 0.738 | 0.637 | |
| FIDProtocol=Reference-based (Top-Ranked Candidate as Reference)2026.01 | 0.624 | 0.525 | |
| HCIIT-JudgeProtocol=No-Reference (Direct Assessment)2026.01 | 0.599 | 0.502 | |
| SSIMProtocol=Reference-based (Top-Ranked Candidate as Reference)2026.01 | 0.514 | 0.412 | |
| PSNRProtocol=Reference-based (Top-Ranked Candidate as Reference)2026.01 | 0.494 | 0.403 | |
| LPIPSProtocol=Reference-based (Top-Ranked Candidate as Reference)2026.01 | 0.299 | 0.24 | |
| SAMScoreProtocol=No-Reference (Direct Assessment)2026.01 | 0.292 | 0.211 | |
| E-Commerce VQAProtocol=No-Reference (Direct Assessment)2026.01 | 0.025 | 0.014 |