9.6 KiB
简体中文 | English
Arcface-Paddle
- 1. Introduction
- 2. Environment Preparation
- 3. Data Preparation
- 4. How to Training
- 5. Model Evaluation
- 6. Export Model
- 7 Model Inference
- 8 Model Performance
- 9. Inference Combined with Face Detection Model
1. Introduction
Arcface-Paddle is an open source deep face detection and recognition toolkit, powered by PaddlePaddle. Arcface-Paddle provides three related pretrained models now, include BlazeFace for face detection, ArcFace and MobileFace for face recognition.
- This tutorial is mainly about face recognition.
- For face detection task, please refer to: Face detection tuturial.
- For Whl package inference using PaddleInference, please refer to whl package inference.
Note: Many thanks to GuoQuanhao for the reproduction of the Arcface basline using PaddlePaddle.
2. Environment Preparation
Please refer to Installation to setup environment at first.
3. Data Preparation
3.1 Download Dataset
Download the dataset from insightface datasets.
- MS1M_v2: MS1M-ArcFace
- MS1M_v3: MS1M-RetinaFace
3.2 Extract MXNet Dataset to images
python tools/mx_recordio_2_images.py --root_dir ms1m-retinaface-t1/ --output_dir MS1M_v3/
After finishing unzipping the dataset, the folder structure is as follows.
MS1M_v3
|_ images
| |_ 00000001.jpg
| |_ ...
| |_ 05179510.jpg
|_ label.txt
|_ agedb_30.bin
|_ cfp_ff.bin
|_ cfp_fp.bin
|_ lfw.bin
Label file format is as follows.
# delimiter: "\t"
# the following the content of label.txt
images/00000001.jpg 0
...
If you want to use customed dataset, you can arrange your data according to the above format.
4. How to Training
4.1 Single Node, Single GPU
export CUDA_VISIBLE_DEVICES=1
python tools/train.py \
--config_file configs/ms1mv2_mobileface.py \
--embedding_size 128 \
--sample_ratio 1.0 \
--loss ArcFace \
--batch_size 512 \
--dataset MS1M_v2 \
--num_classes 85742 \
--data_dir MS1M_v2/ \
--label_file MS1M_v2/label.txt \
--fp16 False
4.2 Single Node, 8 GPUs
Static Mode
sh scripts/train_static.sh
Dynamic Mode
sh scripts/train_dynamic.sh
During training, you can view loss changes in real time through VisualDL, For more information, please refer to VisualDL.
5. Model Evaluation
The model evaluation process can be started as follows.
Static Mode
sh scripts/validation_static.sh
Dynamic Mode
sh scripts/validation_dynamic.sh
6. Export Model
PaddlePaddle supports inference using prediction engines. Firstly, you should export inference model.
Static Mode
sh scripts/export_static.sh
Dynamic Mode
sh scripts/export_dynamic.sh
We also support export to onnx model, you only need to set --export_type onnx.
7. Model Inference
The model inference process supports paddle save inference model and onnx model.
sh scripts/inference.sh
8. Model Performance
8.1 Performance of Lighting Model
Configuration：
- CPU: Intel(R) Xeon(R) Gold 6184 CPU @ 2.40GHz
- GPU: a single NVIDIA Tesla V100
- Precison: FP32
- BatchSize: 64/512
- SampleRatio: 1.0
- Embedding Size: 128
- MS1MV2
| Model structure | lfw | cfp_fp | agedb30 | CPU time cost | GPU time cost | Inference model |
|---|---|---|---|---|---|---|
| MobileFace-Paddle | 0.9952 | 0.9280 | 0.9612 | 4.3ms | 2.3ms | download link |
| MobileFace-mxnet | 0.9950 | 0.8894 | 0.9591 | 7.3ms | 4.7ms | - |
- Note: MobileFace-Paddle training using MobileFaceNet_128
8.2 Accuracy on Verification Datasets
Configuration：
- GPU: 8 NVIDIA Tesla V100 32G
- Precison: Pure FP16
- BatchSize: 128/1024
| Mode | Datasets | backbone | Ratio | agedb30 | cfp_fp | lfw | log | checkpoint |
|---|---|---|---|---|---|---|---|---|
| Static | MS1MV3 | r50 | 0.1 | 0.98317 | 0.98943 | 0.99850 | log | checkpoint |
| Static | MS1MV3 | r50 | 1.0 | 0.98283 | 0.98843 | 0.99850 | log | checkpoint |
| Dynamic | MS1MV3 | r50 | 0.1 | 0.98333 | 0.98900 | 0.99833 | log | checkpoint |
| Dynamic | MS1MV3 | r50 | 1.0 | 0.98317 | 0.98900 | 0.99833 | log | checkpoint |
8.3 Maximum Number of Identities
Configuration：
- GPU: 8 NVIDIA Tesla V100 32G (32510MiB)
- BatchSize: 64/512
- SampleRatio: 0.1
| Mode | Precision | Res50 | Res100 |
|---|---|---|---|
| Framework1 (static) | AMP | 42000000 (31792MiB) | 39000000 (31938MiB) |
| Framework2 (dynamic) | AMP | 30000000 (31702MiB) | 29000000 (32286MiB) |
| Paddle (static) | Pure FP16 | 60000000 (32018MiB) | 60000000 (32018MiB) |
| Paddle (dynamic) | Pure FP16 | 59000000 (31970MiB) | 59000000 (31970MiB) |
Note: config environment variable by export FLAGS_allocator_strategy=naive_best_fit
8.4 Throughtput
Configuration：
- BatchSize: 128/1024
- SampleRatio: 0.1
- Datasets: MS1MV3
- V100: Driver Version: 450.80.02, CUDA Version: 11.0
- A100: Driver Version: 460.32.03, CUDA Version: 11.2
For more experimental results see PLSC, which is an open source Paddle Large Scale Classification Tools powered by PaddlePaddle. It supports 60 million classes on single node 8 NVIDIA V100 (32G).
9. Inference Combined with Face Detection Model
Firstly, use the following commands to download the index gallery, demo image and font file for visualization.
# Index library for the recognition process
wget https://raw.githubusercontent.com/littletomatodonkey/insight-face-paddle/main/demo/friends/index.bin
# Demo image
wget https://raw.githubusercontent.com/littletomatodonkey/insight-face-paddle/main/demo/friends/query/friends2.jpg
# Font file for visualization
wget https://raw.githubusercontent.com/littletomatodonkey/insight-face-paddle/main/SourceHanSansCN-Medium.otf
Use the following command to run the whole face recognition demo.
# detection + recogniotion process
python3.7 tools/test_recognition.py --det --rec --index=index.bin --input=friends2.jpg --output="./output"
The final result is save in folder output/, which is shown as follows.
For more details about parameter explanations, index gallery construction and whl package inference, please refer to: