Image Generation on CELEBA latent dimension 64 (test)
48.71FIDVAE - OURS
Evaluation Results
| Method | Links | |||
|---|---|---|---|---|
| VAE - OURSSampling=Geometry-aware2022.09 | 48.71 | 44 | 62 | |
| HVAE2022.09 | 52 | 38 | 58 | |
| RAELatent Prior=10-component GMM2022.09 | 53.29 | 36 | 58 | |
| RAE (L2)Latent Prior=10-component GMM, Regularization=L22022.09 | 54.45 | 33 | 55 | |
| WAE2022.09 | 54.56 | 57 | 55 | |
| RAE (SN)Latent Prior=10-component GMM, Regularization=Spectral2022.09 | 55.04 | 36 | 56 | |
| RHVAE2022.09 | 55.12 | 45 | 56 | |
| VAE - GMMLatent Prior=10-component GMM2022.09 | 55.5 | 37 | 49 | |
| AE - GMMLatent Prior=10-component GMM2022.09 | 56.07 | 32 | 48 | |
| RAE (GP)Latent Prior=10-component GMM, Regularization=GP2022.09 | 59.41 | 28 | 49 | |
| VAE - N(0, 1)Latent Prior=N(0, 1)2022.09 | 64.13 | 27 | 39 | |
| AE - N(0, 1)Latent Prior=N(0, 1)2022.09 | 64.64 | 29 | 42 | |
| VAMP2022.09 | 73.87 | 9 | 10 |