Image Generation on BreCaHAD 1,750 images (train)
22.61FIDSQ-GAN + ADA
Evaluation results
| Method | Links | |||
|---|---|---|---|---|
| SQ-GAN + ADABase=StyleGAN2, Augmentation=ADA2025.03 | 22.61 | 3.15 | 0.01 | |
| StyleGAN2 + CR + ADARegularization=CR, Augmentation=ADA2025.03 | 22.69 | 2.81 | 0.011 | |
| StyleGAN2 + ADAAugmentation=ADA2025.03 | 23.9 | 3.04 | 0.019 | |
| SQ-GAN + CBIBase=StyleGAN2, Initialization=Knowledge-enhanced codebook initialization2025.03 | 42.42 | 2.93 | 0.037 | |
| SQ-GANBase=StyleGAN22025.03 | 48.44 | 2.98 | 0.041 | |
| StyleGAN2 + CRRegularization=CR2025.03 | 80.72 | 2.92 | 0.058 | |
| StyleGAN2 + LeCamRegularization=LeCam2025.03 | 83.74 | 2.51 | 0.046 | |
| StyleGAN2 + KD-DLGANRegularization=KD-DLGAN2025.03 | 86.25 | 3.26 | 0.083 | |
| StyleGAN22025.03 | 97.06 | 3.1 | 0.095 | |
| StyleGAN2 + DigGANRegularization=DigGAN2025.03 | 105.45 | 3.05 | 0.095 |