Skip to content

Commit 92c6bd9

Browse files
anxiangsiranxiangsir
committed
updated prepare_webface42m.md
1 parent 4df192b commit 92c6bd9

File tree

4 files changed

+39
-63
lines changed

4 files changed

+39
-63
lines changed

‎recognition/arcface_torch/README.md

+17-16
Original file line numberDiff line numberDiff line change
@@ -63,22 +63,23 @@ globalised multi-racial testset contains 242,143 identities and 1,624,305 images
6363

6464
#### 1. Training on Single-Host GPU
6565

66-
| Datasets | Backbone | **MFR-ALL** | IJB-C(1E-4) | IJB-C(1E-5) | log |
67-
|:--------------|:--------------------|:------------|:------------|:------------|:------------------------------------------------------------------------------------------------------------------------------------|
68-
| MS1MV2 | mobilefacenet-0.45G | 62.07 | 93.61 | 90.28 | [click me](https://raw.githubusercontent.com/anxiangsir/insightface_arcface_log/master/ms1mv2_mbf/training.log) |
69-
| MS1MV2 | r50 | 75.13 | 95.97 | 94.07 | [click me](https://raw.githubusercontent.com/anxiangsir/insightface_arcface_log/master/ms1mv2_r50/training.log) |
70-
| MS1MV2 | r100 | 78.12 | 96.37 | 94.27 | [click me](https://raw.githubusercontent.com/anxiangsir/insightface_arcface_log/master/ms1mv2_r100/training.log) |
71-
| MS1MV3 | mobilefacenet-0.45G | 63.78 | 94.23 | 91.33 | [click me](https://raw.githubusercontent.com/anxiangsir/insightface_arcface_log/master/ms1mv3_mbf/training.log) |
72-
| MS1MV3 | r50 | 79.14 | 96.37 | 94.47 | [click me](https://raw.githubusercontent.com/anxiangsir/insightface_arcface_log/master/ms1mv3_r50/training.log) |
73-
| MS1MV3 | r100 | 81.97 | 96.85 | 95.02 | [click me](https://raw.githubusercontent.com/anxiangsir/insightface_arcface_log/master/ms1mv3_r100/training.log) |
74-
| Glint360K | mobilefacenet-0.45G | 70.18 | 95.04 | 92.62 | [click me](https://raw.githubusercontent.com/anxiangsir/insightface_arcface_log/master/glint360k_mbf/training.log) |
75-
| Glint360K | r50 | 86.34 | 97.16 | 95.81 | [click me](https://raw.githubusercontent.com/anxiangsir/insightface_arcface_log/master/glint360k_r50/training.log) |
76-
| Glint360k | r100 | 89.52 | 97.55 | 96.38 | [click me](https://raw.githubusercontent.com/anxiangsir/insightface_arcface_log/master/glint360k_r100/training.log) |
77-
| WF4M | r100 | 89.87 | 97.19 | 95.48 | [click me](https://raw.githubusercontent.com/anxiangsir/insightface_arcface_log/master/wf4m_r100/training.log) |
78-
| WF12M | r100 | 94.69 | 97.59 | 95.97 | [click me](https://raw.githubusercontent.com/anxiangsir/insightface_arcface_log/master/wf12m_r100/training.log) |
79-
| WF12M-PFC-0.2 | r100 | 94.75 | 97.60 | 95.90 | [click me](https://raw.githubusercontent.com/anxiangsir/insightface_arcface_log/master/wf12m_pfc02_r100/training.log) |
80-
| WF42M-PFC-0.2 | R100 | 96.27 | 97.70 | 96.31 | [click me](https://raw.githubusercontent.com/anxiangsir/insightface_arcface_log/master/wf42m_pfc02_r100/training.log) |
81-
| WF42M-PFC-0.2 | ViT-T-1.5G | 92.04 | 97.27 | 95.68 | [click me](https://raw.githubusercontent.com/anxiangsir/insightface_arcface_log/master/wf42m_pfc02_40epoch_8gpu_vit_t/training.log) |
66+
| Datasets | Backbone | **MFR-ALL** | IJB-C(1E-4) | IJB-C(1E-5) | log |
67+
|:---------------|:--------------------|:------------|:------------|:------------|:------------------------------------------------------------------------------------------------------------------------------------|
68+
| MS1MV2 | mobilefacenet-0.45G | 62.07 | 93.61 | 90.28 | [click me](https://raw.githubusercontent.com/anxiangsir/insightface_arcface_log/master/ms1mv2_mbf/training.log) |
69+
| MS1MV2 | r50 | 75.13 | 95.97 | 94.07 | [click me](https://raw.githubusercontent.com/anxiangsir/insightface_arcface_log/master/ms1mv2_r50/training.log) |
70+
| MS1MV2 | r100 | 78.12 | 96.37 | 94.27 | [click me](https://raw.githubusercontent.com/anxiangsir/insightface_arcface_log/master/ms1mv2_r100/training.log) |
71+
| MS1MV3 | mobilefacenet-0.45G | 63.78 | 94.23 | 91.33 | [click me](https://raw.githubusercontent.com/anxiangsir/insightface_arcface_log/master/ms1mv3_mbf/training.log) |
72+
| MS1MV3 | r50 | 79.14 | 96.37 | 94.47 | [click me](https://raw.githubusercontent.com/anxiangsir/insightface_arcface_log/master/ms1mv3_r50/training.log) |
73+
| MS1MV3 | r100 | 81.97 | 96.85 | 95.02 | [click me](https://raw.githubusercontent.com/anxiangsir/insightface_arcface_log/master/ms1mv3_r100/training.log) |
74+
| Glint360K | mobilefacenet-0.45G | 70.18 | 95.04 | 92.62 | [click me](https://raw.githubusercontent.com/anxiangsir/insightface_arcface_log/master/glint360k_mbf/training.log) |
75+
| Glint360K | r50 | 86.34 | 97.16 | 95.81 | [click me](https://raw.githubusercontent.com/anxiangsir/insightface_arcface_log/master/glint360k_r50/training.log) |
76+
| Glint360k | r100 | 89.52 | 97.55 | 96.38 | [click me](https://raw.githubusercontent.com/anxiangsir/insightface_arcface_log/master/glint360k_r100/training.log) |
77+
| WF4M | r100 | 89.87 | 97.19 | 95.48 | [click me](https://raw.githubusercontent.com/anxiangsir/insightface_arcface_log/master/wf4m_r100/training.log) |
78+
| WF12M-PFC-0.2 | r100 | 94.75 | 97.60 | 95.90 | [click me](https://raw.githubusercontent.com/anxiangsir/insightface_arcface_log/master/wf12m_pfc02_r100/training.log) |
79+
| WF12M-PFC-0.3 | r100 | 94.71 | 97.64 | 96.01 | [click me](https://raw.githubusercontent.com/anxiangsir/insightface_arcface_log/master/wf12m_pfc03_r100/training.log) |
80+
| WF12M | r100 | 94.69 | 97.59 | 95.97 | [click me](https://raw.githubusercontent.com/anxiangsir/insightface_arcface_log/master/wf12m_r100/training.log) |
81+
| WF42M-PFC-0.2 | r100 | 96.27 | 97.70 | 96.31 | [click me](https://raw.githubusercontent.com/anxiangsir/insightface_arcface_log/master/wf42m_pfc02_r100/training.log) |
82+
| WF42M-PFC-0.2 | ViT-T-1.5G | 92.04 | 97.27 | 95.68 | [click me](https://raw.githubusercontent.com/anxiangsir/insightface_arcface_log/master/wf42m_pfc02_40epoch_8gpu_vit_t/training.log) |
8283

8384
#### 2. Training on Multi-Host GPU
8485

‎recognition/arcface_torch/docs/eval.md

+12
Original file line numberDiff line numberDiff line change
@@ -24,8 +24,20 @@ CUDA_VISIBLE_DEVICES=0,1 python eval_ijbc.py \
2424
--network iresnet50
2525
```
2626

27+
2728
## Inference
2829

2930
```shell
3031
python inference.py --weight ms1mv3_arcface_r50/backbone.pth --network r50
3132
```
33+
34+
35+
## Result
36+
37+
| Datasets | Backbone | **MFR-ALL** | IJB-C(1E-4) | IJB-C(1E-5) |
38+
|:---------------|:--------------------|:------------|:------------|:------------|
39+
| WF12M-PFC-0.05 | r100 | 94.05 | 97.51 | 95.75 |
40+
| WF12M-PFC-0.1 | r100 | 94.49 | 97.56 | 95.92 |
41+
| WF12M-PFC-0.2 | r100 | 94.75 | 97.60 | 95.90 |
42+
| WF12M-PFC-0.3 | r100 | 94.71 | 97.64 | 96.01 |
43+
| WF12M | r100 | 94.69 | 97.59 | 95.97 |

‎recognition/arcface_torch/docs/install.md

+5-45
Original file line numberDiff line numberDiff line change
@@ -1,51 +1,11 @@
1-
## v1.8.0
2-
### Linux and Windows
3-
```shell
4-
# CUDA 11.0
5-
pip --default-timeout=100 install torch==1.8.0+cu111 torchvision==0.9.0+cu111 torchaudio==0.8.0 -f https://download.pytorch.org/whl/torch_stable.html
6-
7-
# CUDA 10.2
8-
pip --default-timeout=100 install torch==1.8.0 torchvision==0.9.0 torchaudio==0.8.0
9-
10-
# CPU only
11-
pip --default-timeout=100 install torch==1.8.0+cpu torchvision==0.9.0+cpu torchaudio==0.8.0 -f https://download.pytorch.org/whl/torch_stable.html
1+
## [v1.11.0](https://pytorch.org/)
122

13-
```
14-
15-
16-
## v1.7.1
3+
## [v1.9.0](https://pytorch.org/get-started/previous-versions/#linux-and-windows-7)
174
### Linux and Windows
185
```shell
19-
# CUDA 11.0
20-
pip install torch==1.7.1+cu110 torchvision==0.8.2+cu110 torchaudio==0.7.2 -f https://download.pytorch.org/whl/torch_stable.html
6+
# CUDA 11.1
7+
pip install torch==1.9.0+cu111 torchvision==0.10.0+cu111 torchaudio==0.9.0 -f https://download.pytorch.org/whl/torch_stable.html
218

229
# CUDA 10.2
23-
pip install torch==1.7.1 torchvision==0.8.2 torchaudio==0.7.2
24-
25-
# CUDA 10.1
26-
pip install torch==1.7.1+cu101 torchvision==0.8.2+cu101 torchaudio==0.7.2 -f https://download.pytorch.org/whl/torch_stable.html
27-
28-
# CUDA 9.2
29-
pip install torch==1.7.1+cu92 torchvision==0.8.2+cu92 torchaudio==0.7.2 -f https://download.pytorch.org/whl/torch_stable.html
30-
31-
# CPU only
32-
pip install torch==1.7.1+cpu torchvision==0.8.2+cpu torchaudio==0.7.2 -f https://download.pytorch.org/whl/torch_stable.html
10+
pip install torch==1.9.0+cu102 torchvision==0.10.0+cu102 torchaudio==0.9.0 -f https://download.pytorch.org/whl/torch_stable.html
3311
```
34-
35-
36-
## v1.6.0
37-
38-
### Linux and Windows
39-
```shell
40-
# CUDA 10.2
41-
pip install torch==1.6.0 torchvision==0.7.0
42-
43-
# CUDA 10.1
44-
pip install torch==1.6.0+cu101 torchvision==0.7.0+cu101 -f https://download.pytorch.org/whl/torch_stable.html
45-
46-
# CUDA 9.2
47-
pip install torch==1.6.0+cu92 torchvision==0.7.0+cu92 -f https://download.pytorch.org/whl/torch_stable.html
48-
49-
# CPU only
50-
pip install torch==1.6.0+cpu torchvision==0.7.0+cpu -f https://download.pytorch.org/whl/torch_stable.html
51-
```

‎recognition/arcface_torch/docs/prepare_webface42m.md

+5-2
Original file line numberDiff line numberDiff line change
@@ -3,8 +3,11 @@
33

44
## 1. Download Datasets and Unzip
55

6-
Download WebFace42M from [https://www.face-benchmark.org/download.html](https://www.face-benchmark.org/download.html).
7-
6+
Download WebFace42M from [https://www.face-benchmark.org/download.html](https://www.face-benchmark.org/download.html).
7+
The raw data of `WebFace42M` will have 10 directories after being unarchived:
8+
`WebFace4M` contains 1 directory: `0`.
9+
`WebFace12M` contains 3 directories: `0,1,2`.
10+
`WebFace42M` contains 10 directories: `0,1,2,3,4,5,6,7,8,9`.
811

912
## 2. Create Shuffled Rec File for DALI
1013

0 commit comments

Comments
 (0)