https://github.com/yas-sim/dbface-on-openvino
Describes how to run DBFace, a real-time, single-shot face detection model on Intel OpenVINO
https://github.com/yas-sim/dbface-on-openvino
deep-learning face-detection inference-engine intel onnx openvino python pytorch sample
Last synced: 2 months ago
JSON representation
Describes how to run DBFace, a real-time, single-shot face detection model on Intel OpenVINO
- Host: GitHub
- URL: https://github.com/yas-sim/dbface-on-openvino
- Owner: yas-sim
- License: apache-2.0
- Created: 2020-06-27T06:42:01.000Z (over 5 years ago)
- Default Branch: master
- Last Pushed: 2020-08-23T23:20:20.000Z (about 5 years ago)
- Last Synced: 2025-03-29T15:34:54.222Z (6 months ago)
- Topics: deep-learning, face-detection, inference-engine, intel, onnx, openvino, python, pytorch, sample
- Language: Jupyter Notebook
- Homepage:
- Size: 16.3 MB
- Stars: 29
- Watchers: 1
- Forks: 7
- Open Issues: 3
-
Metadata Files:
- Readme: README.md
- License: LICENSE
Awesome Lists containing this project
README
# Overview
The objective of this project is to run the DBFace, a real-time, single-stage face detector on Intel(r) Distribution of OpenVINO(tm) Toolkit.
**Original DBFace GitHub site:**
https://github.com/dlunion/DBFace
Thie original developer used PyTorch to train the model. You need to convert the PyTorch model to OpenVINO IR model. You need to run 2 conversion steps to get the final IR model.
```sh
PyTorch (.pth) -> ONNX (.onnx) -> IR (.xml, .bin)
```
このプロジェクトの目的はリアルタイムでシングルステージの顔検出モデル、DBFaceをIntel(r) Distribution of OpenVINO(tm) Toolkit上で実行できるようにすることです。
**Original DBFace GitHub site:**
https://github.com/dlunion/DBFace
元の開発者はPyTorchを使ってモデルの学習を行っています。そのためPyTorchモデルからOpenVINO IRモデルへの変換を行わなくてはなりません。 IRモデルを得るためには２つの変換ステップを要します。
```sh
PyTorch (.pth) -> ONNX (.onnx) -> IR (.xml, .bin)
```
**Detection result with a 4VGA (1280x960) input picture**
## 1. Prerequisites
* [Intel Distribution of OpenVINO toolkit 2020.2 (or newer)](https://software.intel.com/content/www/us/en/develop/tools/openvino-toolkit.html)
* Model Downloader / Model Converter setup
[Model Downloader guide](https://docs.openvinotoolkit.org/latest/_tools_downloader_README.html)
```sh
pip3 -r ${INTEL_OPENVINO_DIR}/deployment_tools/tools/model_downloader/requirements.in
pip3 -r ${INTEL_OPENVINO_DIR}/deployment_tools/tools/model_downloader/requirements-pytorch.in
```
If you fail to install PyTorch, go to PyTorch official web site and follow the `QUICK START LOCALLY` guide to install it. You need to have PyTorch >=1.4.
## 2. Download DBFace PyTorch model and weight
Download `DBFace.py` and `dbface.pth` from [original developer's GitHub page](https://github.com/dlunion/DBFace/tree/master/model).
## 3. Convert PyTorch model into ONNX model
Use `pytorch_to_onnx.py` in the `model_downloader` directory.
```sh
python3 ${INTEL_OPENVINO_DIR}/deployment_tools/tools/model_downloader/pytorch_to_onnx.py \
--model-name DBFace \
--weights dbface.pth \
--import-module DBFace \
--input-shape 1,3,320,320 \
--output-file dbface.onnx \
--input-names x \
--output-names sigmoid_hm,tlrb,landmark
```
**Hint:** You can get the input and output node names from the model source code (in this case, `DBFace.py` in the `model` directory)
```Python
def forward(self, x):
out = self.hs1(self.bn1(self.conv1(x)))
:
return sigmoid_hm, tlrb, landmark
```
## 4. Convert ONNX model into OpenVINO IR model
Use `Model Optimizer (MO)` to convert the ONNX model into IR model.
```sh
python3 ${INTEL_OPENVINO_DIR}/deployment_tools/model_optimizer/mo.py \
--input_model dbface.onnx \
--mean_values [180,154,150] \
--scale_values [73.7,70.0,70.9] \
--input_shape [1,3,960,1280] \
--output_dir dbface-4vga \
--data_type FP16
```
**Hint:** You can change the input shape with `--input_shape` option.
**Hint:** You can find the appropriate parameters (`--mean_values` and `--scale_values`) for input data preprocessing from the original source code (in this case, line 36-40 in `main.py` from the origianl GitHub site)
```Python
mean = [0.408, 0.447, 0.47]
std = [0.289, 0.274, 0.278]
image = common.pad(image)
image = ((image / 255.0 - mean) / std).astype(np.float32)
```
## 5. Run sample program
```sh
python3 dbface-infer.py -m model.xml -i input_image
```
A webCam #0 will be used when you specify 'cam' as the input file name.
`output.jpg` will be ganarated in the current directry. (non-webCam version only)
**Command line example:**
```sh
$ python3 dbface-infer.py -m dbface-4vga/dbface.xml -i image.jpg
$ python3 dbface-infer.py -m dbface-vga/dbface.xml -i cam
```
## 6. Test Environment
- Ubuntu 18.04 / Windows 10 1909
- OpenVINO 2020.3 LTS
## See Also
* [GitHub, dlunion/DBFace](https://github.com/dlunion/DBFace)
* [Intel(r) Distribution of OpenVINO(tm) Toolkit](https://software.intel.com/content/www/us/en/develop/tools/openvino-toolkit.html)
* [Converting a ONNX* Model](https://docs.openvinotoolkit.org/latest/_docs_MO_DG_prepare_model_convert_model_Convert_Model_From_ONNX.html)
* [Model Downloader and other automation tools](https://docs.openvinotoolkit.org/latest/_tools_downloader_README.html)
* [PyTorch official web page](https://pytorch.org/?utm_source=Google&utm_medium=PaidSearch&utm_campaign=%2A%2ALP+-+TM+-+General+-+HV+-+JP&utm_adgroup=Installing+PyTorch&utm_keyword=installing%20pytorch&utm_offering=AI&utm_Product=PyTorch&gclid=Cj0KCQjwudb3BRC9ARIsAEa-vUvjBVIyGnP31gCk__x1bquCw5HNX3Av0Mu0vwU75HBxgT79lCdwsuUaAoUFEALw_wcB)