English | 简体中文
- InsightFace
- （1）The *.pt provided by the Official Library can be deployed after the Export ONNX Model；
- （2）As for InsightFace model trained on customized data, please follow the operations guidelines in Export ONNX Model to complete the deployment.
Now FastDeploy supports the deployment of the following models
- ArcFace
- CosFace
- PartialFC
- VPL
Taking ArcFace as an example:
Visit ArcFace official github repository, follow the guidelines to download pt model files, and employ torch2onnx.py to get the file in onnx format.
-
Download ArcFace model files
Link: https://pan.baidu.com/share/init?surl=CL-l4zWqsI1oDuEEYVhj-g code: e8pw -
Export files in onnx format
PYTHONPATH=. python ./torch2onnx.py ms1mv3_arcface_r100_fp16/backbone.pth --output ms1mv3_arcface_r100.onnx --network r100 --simplify 1
For developers' testing, models exported by InsightFace are provided below. Developers can download and use them directly. (The accuracy of the models in the table is sourced from the official library) The accuracy metric is sourced from the model description in InsightFace. Refer to the introduction in InsightFace for more details.
| Model | Size | Accuracy (AgeDB_30) |
|---|---|---|
| CosFace-r18 | 92MB | 97.7 |
| CosFace-r34 | 131MB | 98.3 |
| CosFace-r50 | 167MB | 98.3 |
| CosFace-r100 | 249MB | 98.4 |
| ArcFace-r18 | 92MB | 97.7 |
| ArcFace-r34 | 131MB | 98.1 |
| ArcFace-r50 | 167MB | - |
| ArcFace-r100 | 249MB | 98.4 |
| ArcFace-r100_lr0.1 | 249MB | 98.4 |
| PartialFC-r34 | 167MB | - |
| PartialFC-r50 | 249MB | - |
- This tutorial and related code are written based on InsightFace CommitID:babb9a5