Spaces:
Running
on
Zero
Running
on
Zero
Add files
Browse files- .gitmodules +3 -0
- app.py +142 -0
- face_detection +1 -0
- requirements.txt +4 -0
.gitmodules
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
[submodule "face_detection"]
|
2 |
+
path = face_detection
|
3 |
+
url = https://github.com/ibug-group/face_detection
|
app.py
ADDED
@@ -0,0 +1,142 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
#!/usr/bin/env python
|
2 |
+
|
3 |
+
from __future__ import annotations
|
4 |
+
|
5 |
+
import argparse
|
6 |
+
import functools
|
7 |
+
import pathlib
|
8 |
+
import sys
|
9 |
+
import urllib.request
|
10 |
+
from typing import Union
|
11 |
+
|
12 |
+
import cv2
|
13 |
+
import gradio as gr
|
14 |
+
import numpy as np
|
15 |
+
import torch
|
16 |
+
import torch.nn as nn
|
17 |
+
|
18 |
+
sys.path.insert(0, 'face_detection')
|
19 |
+
|
20 |
+
from ibug.face_detection import RetinaFacePredictor, S3FDPredictor
|
21 |
+
|
22 |
+
REPO_URL = 'https://github.com/ibug-group/face_detection'
|
23 |
+
TITLE = 'ibug-group/face_detection'
|
24 |
+
DESCRIPTION = f'This is a demo for {REPO_URL}.'
|
25 |
+
ARTICLE = None
|
26 |
+
|
27 |
+
|
28 |
+
def parse_args() -> argparse.Namespace:
|
29 |
+
parser = argparse.ArgumentParser()
|
30 |
+
parser.add_argument('--face-score-slider-step', type=float, default=0.05)
|
31 |
+
parser.add_argument('--face-score-threshold', type=float, default=0.8)
|
32 |
+
parser.add_argument('--device', type=str, default='cpu')
|
33 |
+
parser.add_argument('--theme', type=str)
|
34 |
+
parser.add_argument('--live', action='store_true')
|
35 |
+
parser.add_argument('--share', action='store_true')
|
36 |
+
parser.add_argument('--port', type=int)
|
37 |
+
parser.add_argument('--disable-queue',
|
38 |
+
dest='enable_queue',
|
39 |
+
action='store_false')
|
40 |
+
parser.add_argument('--allow-flagging', type=str, default='never')
|
41 |
+
parser.add_argument('--allow-screenshot', action='store_true')
|
42 |
+
return parser.parse_args()
|
43 |
+
|
44 |
+
|
45 |
+
def load_model(
|
46 |
+
model_name: str, threshold: float,
|
47 |
+
device: torch.device) -> Union[RetinaFacePredictor, S3FDPredictor]:
|
48 |
+
if model_name == 's3fd':
|
49 |
+
model = S3FDPredictor(threshold=threshold, device=device)
|
50 |
+
else:
|
51 |
+
model_name = model_name.replace('retinaface_', '')
|
52 |
+
model = RetinaFacePredictor(
|
53 |
+
threshold=threshold,
|
54 |
+
device=device,
|
55 |
+
model=RetinaFacePredictor.get_model(model_name))
|
56 |
+
return model
|
57 |
+
|
58 |
+
|
59 |
+
def detect(image: np.ndarray, model_name: str, face_score_threshold: float,
|
60 |
+
detectors: dict[str, nn.Module]) -> np.ndarray:
|
61 |
+
model = detectors[model_name]
|
62 |
+
model.threshold = face_score_threshold
|
63 |
+
|
64 |
+
# RGB -> BGR
|
65 |
+
image = image[:, :, ::-1]
|
66 |
+
preds = model(image, rgb=False)
|
67 |
+
|
68 |
+
res = image.copy()
|
69 |
+
for pred in preds:
|
70 |
+
box = np.round(pred[:4]).astype(int)
|
71 |
+
|
72 |
+
line_width = max(2, int(3 * (box[2:] - box[:2]).max() / 256))
|
73 |
+
cv2.rectangle(res, tuple(box[:2]), tuple(box[2:]), (0, 255, 0),
|
74 |
+
line_width)
|
75 |
+
|
76 |
+
if len(pred) == 15:
|
77 |
+
pts = pred[5:].reshape(-1, 2)
|
78 |
+
for pt in np.round(pts).astype(int):
|
79 |
+
cv2.circle(res, tuple(pt), line_width, (0, 255, 0), cv2.FILLED)
|
80 |
+
|
81 |
+
return res[:, :, ::-1]
|
82 |
+
|
83 |
+
|
84 |
+
def main():
|
85 |
+
gr.close_all()
|
86 |
+
|
87 |
+
args = parse_args()
|
88 |
+
device = torch.device(args.device)
|
89 |
+
|
90 |
+
model_names = [
|
91 |
+
'retinaface_mobilenet0.25',
|
92 |
+
'retinaface_resnet50',
|
93 |
+
's3fd',
|
94 |
+
]
|
95 |
+
detectors = {
|
96 |
+
name: load_model(name,
|
97 |
+
threshold=args.face_score_threshold,
|
98 |
+
device=device)
|
99 |
+
for name in model_names
|
100 |
+
}
|
101 |
+
|
102 |
+
func = functools.partial(detect, detectors=detectors)
|
103 |
+
func = functools.update_wrapper(func, detect)
|
104 |
+
|
105 |
+
image_path = pathlib.Path('selfie.jpg')
|
106 |
+
if not image_path.exists():
|
107 |
+
url = 'https://raw.githubusercontent.com/peiyunh/tiny/master/data/demo/selfie.jpg'
|
108 |
+
urllib.request.urlretrieve(url, image_path)
|
109 |
+
examples = [[image_path.as_posix(), model_names[1], 0.8]]
|
110 |
+
|
111 |
+
gr.Interface(
|
112 |
+
func,
|
113 |
+
[
|
114 |
+
gr.inputs.Image(type='numpy', label='Input'),
|
115 |
+
gr.inputs.Radio(model_names,
|
116 |
+
type='value',
|
117 |
+
default='retinaface_resnet50',
|
118 |
+
label='Model'),
|
119 |
+
gr.inputs.Slider(0,
|
120 |
+
1,
|
121 |
+
step=args.face_score_slider_step,
|
122 |
+
default=args.face_score_threshold,
|
123 |
+
label='Face Score Threshold'),
|
124 |
+
],
|
125 |
+
gr.outputs.Image(type='numpy', label='Output'),
|
126 |
+
examples=examples,
|
127 |
+
title=TITLE,
|
128 |
+
description=DESCRIPTION,
|
129 |
+
article=ARTICLE,
|
130 |
+
theme=args.theme,
|
131 |
+
allow_screenshot=args.allow_screenshot,
|
132 |
+
allow_flagging=args.allow_flagging,
|
133 |
+
live=args.live,
|
134 |
+
).launch(
|
135 |
+
enable_queue=args.enable_queue,
|
136 |
+
server_port=args.port,
|
137 |
+
share=args.share,
|
138 |
+
)
|
139 |
+
|
140 |
+
|
141 |
+
if __name__ == '__main__':
|
142 |
+
main()
|
face_detection
ADDED
@@ -0,0 +1 @@
|
|
|
|
|
1 |
+
Subproject commit bc1e392b11d731fa20b1397c8ff3faed5e7fc76e
|
requirements.txt
ADDED
@@ -0,0 +1,4 @@
|
|
|
|
|
|
|
|
|
|
|
1 |
+
numpy==1.22.3
|
2 |
+
opencv-python-headless==4.5.5.64
|
3 |
+
torch==1.11.0
|
4 |
+
torchvision==0.12.0
|