Update README.md
Browse files
README.md
CHANGED
|
@@ -1,3 +1,19 @@
|
|
| 1 |
-
---
|
| 2 |
-
license: apache-2.0
|
| 3 |
-
---
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
---
|
| 2 |
+
license: apache-2.0
|
| 3 |
+
---
|
| 4 |
+
|
| 5 |
+
### Hierarchical Variational Autoencoder (HVAE)
|
| 6 |
+
|
| 7 |
+
Generative model for 2D B-mode Ultrasound Images.
|
| 8 |
+
This model can be used for unconditional sampling as well as inpainting tasks.
|
| 9 |
+
|
| 10 |
+
Current architecture on at resolution `256x256x3` (3 video frames per input/output).
|
| 11 |
+
|
| 12 |
+
🌽 The `zea` implementation is a snippet, adapted from the [source](https://github.com/swpenninga/hvae).
|
| 13 |
+
|
| 14 |
+
The weights that are currently present can be read as follows:
|
| 15 |
+
- `lvh`: trained on `EchonetLVH` dataset
|
| 16 |
+
- `ur`: encoder retrained with UniformRandom subsampling (`zea.agent`)
|
| 17 |
+
- `24`: 24 lines/columns used when subsampling
|
| 18 |
+
|
| 19 |
+
📚 A usage tutorial can be found at:
|