File size: 2,224 Bytes
f711495 c3c0c9e f711495 e275878 da99f1f c3c0c9e e275878 f711495 e275878 f711495 e275878 ea0fcb6 e275878 8849c8a f5aae88 ea0fcb6 e275878 008a295 e275878 ea7cb42 e275878 c3c0c9e |
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 |
---
license: apache-2.0
language:
- en
pipeline_tag: image-to-image
tags:
- art
- lama
- inpainting
---
# LaMa Inpainting Model
This ONNX model is a port of the original PyTorch big-lama model.
> HG Space: https://huggingface.co/spaces/Carve/LaMa-Demo-ONNX
## Description
There are two versions of the model:
### 1. `lama_fp32.onnx` (RECOMMENDED)
This version was exported using the old torch to ONNX converter (`torch.onnx.export`).
**Notes:**
1. **Custom FourierUnitJIT**: A custom [FourierUnitJIT](https://github.com/Carve-Photos/lama/blob/main/saicinpainting/training/modules/ffc.py) implementation is used since the original cannot be directly ported to ONNX without overhead. The result is identical to the original model.
2. **Fixed Input Shape**: The input shape is fixed at 512x512 pixels. Although dynamic input shapes are possible, they would require resolving issues with dynamic padding in the `irfft` and `rfftn` functions in `ffc.py`.
3. **Opset Version 17**: This model uses opset version 17.
4. **Exportable to TensorRT**: The model can be successfully used in TensorRT, etc.
> if you need other resolution - export it using our [jupyter notebook](https://colab.research.google.com/github/Carve-Photos/lama/blob/main/export_LaMa_to_onnx.ipynb)
### 2. `lama.onnx` (NOT RECOMMENDED)
This version was exported using the new torch to ONNX converter (`torch.onnx.dynamo_export`).
**Notes:**
1. **Custom DFT irfftn Logic**: Uses a custom irfftn ONNX logic (patched `onnxscript`).
2. **Fixed Input Shape**: The input shape is fixed at 512x512 pixels.
3. **Opset Version 18**: This model uses opset version 18.
4. **Performance**: The model works slowly due to issues with `torch.onnx.dynamo_export` and optimization of the ONNX model.
## Resources
- Original repository: [advimman/lama](https://github.com/advimman/lama)
- Repository with custom implementation of exportable LaMa: [Carve-Photos/lama](https://github.com/Carve-Photos/lama)
## Example
**Original image:**
**lama_fp32.onnx - output:**
**lama.onnx - output:**
**Original model output:**
 |