Text To Image Generation On Multi Modal
평가 지표
Acc
FID
LPIPS
Real
평가 결과
이 벤치마크에서 각 모델의 성능 결과
비교 표
| 모델 이름 | Acc | FID | LPIPS | Real |
|---|---|---|---|---|
| dm-gan-dynamic-memory-generative-adversarial | 16.4 | 131.05 | 0.544 | 16.9 |
| shifted-diffusion-for-text-to-image | - | 19.74 | - | - |
| tedigan-text-guided-diverse-image-generation | 18.4 | 106.37 | 0.456 | 22.6 |
| df-gan-deep-fusion-generative-adversarial | 17.3 | 137.60 | 0.581 | 14.5 |
| lafite-towards-language-free-training-for | - | 12.54 | - | - |
| controllable-text-to-image-generation | 14.6 | 116.32 | 0.522 | 13.1 |
| towards-open-world-text-guided-face-image | 20.4 | 101.42 | 0.461 | 21.0 |
| attngan-fine-grained-text-to-image-generation | 13.0 | 125.98 | 0.512 | 11.9 |
| swinv2-imagen-hierarchical-vision-transformer | - | 10.31 | - | - |
| unite-and-conquer-cross-dataset-multimodal | - | 26.09 | 0.519 | - |