Update README.md
Browse files
README.md
CHANGED
|
@@ -2,12 +2,13 @@
|
|
| 2 |
license: apache-2.0
|
| 3 |
---
|
| 4 |
# LaMa Inpainting Model
|
| 5 |
-
This is an ONNX model ported from original PyTorch big-lama model.
|
| 6 |
|
| 7 |
## Notes:
|
| 8 |
1. We use a custom FourierUnit implementation, as the original cannot be directly ported to ONNX without overhead. The result is not the same, but very close
|
| 9 |
2. The input shape is fixed to 512x512 pixels.
|
| 10 |
3. The opset version is 18.
|
|
|
|
| 11 |
|
| 12 |
|
| 13 |
Original repo: https://github.com/advimman/lama
|
|
|
|
| 2 |
license: apache-2.0
|
| 3 |
---
|
| 4 |
# LaMa Inpainting Model
|
| 5 |
+
This is an alpha version of ONNX model ported from original PyTorch big-lama model.
|
| 6 |
|
| 7 |
## Notes:
|
| 8 |
1. We use a custom FourierUnit implementation, as the original cannot be directly ported to ONNX without overhead. The result is not the same, but very close
|
| 9 |
2. The input shape is fixed to 512x512 pixels.
|
| 10 |
3. The opset version is 18.
|
| 11 |
+
4. The model works slowly due to issues with torch.onnx.dynamo_export with optimization of the onnx model.
|
| 12 |
|
| 13 |
|
| 14 |
Original repo: https://github.com/advimman/lama
|