Update README.md
Browse files
README.md
CHANGED
|
@@ -5,7 +5,7 @@ datasets:
|
|
| 5 |
|
| 6 |
| Generated | Real (for comparison) |
|
| 7 |
| ----- | --------- |
|
| 8 |
-
|  dataset. The model uses [Progressive Growing](https://arxiv.org/pdf/1710.10196.pdf) with [Spectral Normalization](https://arxiv.org/pdf/1802.05957.pdf).
|
| 11 |
|
|
@@ -21,13 +21,20 @@ A significant improvement over https://huggingface.co/PrakhAI/AIPlane2 is the el
|
|
| 21 |
| - | - |
|
| 22 |
|  |  |
|
| 23 |
|
| 24 |
-
# Samples
|
| 25 |
-
|
| 26 |
|
|
|
|
| 27 |
|
| 28 |
# ProGAN
|
| 29 |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 30 |
|
|
|
|
| 31 |
|
| 32 |
# Spectral Normalization
|
| 33 |
|
|
|
|
| 5 |
|
| 6 |
| Generated | Real (for comparison) |
|
| 7 |
| ----- | --------- |
|
| 8 |
+
|  |  |
|
| 9 |
|
| 10 |
This GAN model is trained on the [FGVC Aircraft](https://www.robots.ox.ac.uk/~vgg/data/fgvc-aircraft/) dataset. The model uses [Progressive Growing](https://arxiv.org/pdf/1710.10196.pdf) with [Spectral Normalization](https://arxiv.org/pdf/1802.05957.pdf).
|
| 11 |
|
|
|
|
| 21 |
| - | - |
|
| 22 |
|  |  |
|
| 23 |
|
| 24 |
+
# 'Good' Generated Samples
|
|
|
|
| 25 |
|
| 26 |
+
|
| 27 |
|
| 28 |
# ProGAN
|
| 29 |
|
| 30 |
+
Progressive Growing of GANs was proposed in [Progressive Growing of GANs for improved Quality, Stability, and Variation](https://arxiv.org/pdf/1710.10196.pdf)
|
| 31 |
+
|
| 32 |
+
The idea is to start learning at lower resolutions, and growing the resolution of the GAN over time. This improves both:
|
| 33 |
+
|
| 34 |
+
- Training Speed: At lower resolutions, the Generator and Discriminator have fewer layers and fewer parameters.
|
| 35 |
+
- Convergence Speed: It is much easier to learn high-level details followed by higher granularity features, compared to learning both at the same time.
|
| 36 |
|
| 37 |
+
|
| 38 |
|
| 39 |
# Spectral Normalization
|
| 40 |
|