Few-shot Font Generation on Large dataset (Unseen Fonts Seen Characters (UFSC))
0.2358RMSEGAR-Font(M4)
Evaluation Results
| Method | Links | ||||||
|---|---|---|---|---|---|---|---|
| GAR-Font(M4)model variant=M4, post-training=NFA and SE2026.01 | 0.2358 | 0.6712 | 0.0796 | 8.8073 | 98 | 45.66 | |
| GAR-Font(M2)model variant=M2, post-training=NFA and SE2026.01 | 0.2361 | 0.6707 | 0.0799 | 8.8867 | 98.17 | 45.08 | |
| Vision-only Baselinenumber of reference images (n_ref)=8, post-training=NFA and SE2026.01 | 0.2398 | 0.6627 | 0.0831 | 8.3751 | 97.76 | 41.54 | |
| Vision-only Baselinenumber of reference images (n_ref)=4, post-training=NFA and SE2026.01 | 0.2437 | 0.6552 | 0.0848 | 9.596 | 98.39 | 37.11 | |
| Vision-only Baselinenumber of reference images (n_ref)=2, post-training=NFA and SE2026.01 | 0.2501 | 0.6438 | 0.0888 | 10.1464 | 98.51 | 33.14 |