In-Image Machine Translation Evaluation on Vectra-Bench zh→fr 1.0
0.708Pearson CorrelationVectra
Evaluation results
| Method | Links | ||
|---|---|---|---|
| VectraProtocol=No-Reference (Direct Assessment)2026.01 | 0.708 | 0.613 | |
| HCIIT-JudgeProtocol=No-Reference (Direct Assessment)2026.01 | 0.627 | 0.541 | |
| FIDProtocol=Reference-based (Top-Ranked Candidate as Reference)2026.01 | 0.618 | 0.516 | |
| SSIMProtocol=Reference-based (Top-Ranked Candidate as Reference)2026.01 | 0.447 | 0.359 | |
| PSNRProtocol=Reference-based (Top-Ranked Candidate as Reference)2026.01 | 0.443 | 0.364 | |
| LPIPSProtocol=Reference-based (Top-Ranked Candidate as Reference)2026.01 | 0.262 | 0.215 | |
| SAMScoreProtocol=No-Reference (Direct Assessment)2026.01 | 0.234 | 0.166 | |
| E-Commerce VQAProtocol=No-Reference (Direct Assessment)2026.01 | 0.093 | 0.074 |