Create README.md
Browse files
README.md
ADDED
|
@@ -0,0 +1,27 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
|
| 2 |
+
## EQ-VAE: Equivariance Regularized Latent Space for Improved Generative Image Modeling
|
| 3 |
+
Arxiv: https://arxiv.org/abs/2502.09509 <br>
|
| 4 |
+
|
| 5 |
+
**EQ-VAE** regularizes the latent space of pretrained autoencoders by enforcing equivariance under scaling and rotation transformations.
|
| 6 |
+
|
| 7 |
+
---
|
| 8 |
+
#### Model Description
|
| 9 |
+
This model is a regularized version of [SD-VAE](https://github.com/CompVis/latent-diffusion). We finetune it with EQ-VAE regularization for 44 epochs on Imagenet with EMA weights.
|
| 10 |
+
|
| 11 |
+
|
| 12 |
+
## Model Usage
|
| 13 |
+
These weights are intended to be used with the [EQ-VAE codebase](https://github.com/zelaki/eqvae) or the [CompVis Stable Diffusion codebase](https://github.com/CompVis/stable-diffusion).
|
| 14 |
+
If you are looking for the model to use with the 🧨 diffusers library, [come here](https://huggingface.co/zelaki/eq-vae-ema).
|
| 15 |
+
|
| 16 |
+
|
| 17 |
+
#### Metrics
|
| 18 |
+
Reconstruction performance of eq-vae-ema on Imagenet Validation Set.
|
| 19 |
+
|
| 20 |
+
| **Metric** | **Score** |
|
| 21 |
+
|------------|-----------|
|
| 22 |
+
| **FID** | 0.552 |
|
| 23 |
+
| **PSNR** | 26.158 |
|
| 24 |
+
| **LPIPS** | 0.133 |
|
| 25 |
+
| **SSIM** | 0.725 |
|
| 26 |
+
---
|
| 27 |
+
|