In-Image Machine Translation Evaluation on Vectra-Bench zh→en 1.0
0.8Pearson Correlation (r)Vectra
Evaluation Results
| Method | Links | ||
|---|---|---|---|
| VectraProtocol=No-Reference (Direct Assessment)2026.01 | 0.8 | 0.709 | |
| FIDProtocol=Reference-based (Top-Ranked Candidate as Reference)2026.01 | 0.66 | 0.557 | |
| HCIIT-JudgeProtocol=No-Reference (Direct Assessment)2026.01 | 0.602 | 0.496 | |
| SSIMProtocol=Reference-based (Top-Ranked Candidate as Reference)2026.01 | 0.567 | 0.448 | |
| PSNRProtocol=Reference-based (Top-Ranked Candidate as Reference)2026.01 | 0.537 | 0.434 | |
| SAMScoreProtocol=No-Reference (Direct Assessment)2026.01 | 0.337 | 0.231 | |
| LPIPSProtocol=Reference-based (Top-Ranked Candidate as Reference)2026.01 | 0.303 | 0.237 | |
| E-Commerce VQAProtocol=No-Reference (Direct Assessment)2026.01 | 0.047 | 0.031 |