Image Generation on SVHN latent dimension 16 (test)
46.99FIDVAE - OURS
Evaluation Results
| Method | Links | |||
|---|---|---|---|---|
| VAE - OURSSampling=Geometry-aware2022.09 | 46.99 | 84 | 85 | |
| WAE2022.09 | 49.07 | 80 | 85 | |
| VAE - GMMLatent Prior=10-component GMM2022.09 | 52.32 | 82 | 85 | |
| AE - GMMLatent Prior=10-component GMM2022.09 | 54.21 | 82 | 83 | |
| RAE (L2)Latent Prior=10-component GMM, Regularization=L22022.09 | 58.32 | 82 | 79 | |
| RAE (GP)Latent Prior=10-component GMM, Regularization=GP2022.09 | 61.43 | 79 | 78 | |
| RAELatent Prior=10-component GMM2022.09 | 66.2 | 76 | 80 | |
| VAE - N(0, 1)Latent Prior=N(0, 1)2022.09 | 83.55 | 69 | 55 | |
| VAMP2022.09 | 91.98 | 55 | 63 | |
| RAE (SN)Latent Prior=10-component GMM, Regularization=Spectral2022.09 | 95.64 | 53 | 63 | |
| HVAE2022.09 | 98.05 | 64 | 68 | |
| AE - N(0, 1)Latent Prior=N(0, 1)2022.09 | 119.65 | 54 | 37 | |
| RHVAE2022.09 | 121.69 | 55 | 41 |