Christen Millerdurai
commited on
Commit
•
9564652
1
Parent(s):
15bc41b
init
Browse files- .gitignore +2 -1
- app.py +150 -15
- demo.py → infererence.py +45 -96
.gitignore
CHANGED
@@ -159,4 +159,5 @@ cython_debug/
|
|
159 |
# option (not recommended) you can uncomment the following to ignore the entire idea folder.
|
160 |
#.idea/
|
161 |
src/Ev2Hands/outputs
|
162 |
-
src/HandSimulator/logs
|
|
|
|
159 |
# option (not recommended) you can uncomment the following to ignore the entire idea folder.
|
160 |
#.idea/
|
161 |
src/Ev2Hands/outputs
|
162 |
+
src/HandSimulator/logs
|
163 |
+
gradio_cached_examples/
|
app.py
CHANGED
@@ -1,25 +1,160 @@
|
|
1 |
import gradio as gr
|
2 |
-
import
|
|
|
|
|
3 |
|
|
|
|
|
|
|
4 |
|
5 |
|
6 |
-
|
7 |
-
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
8 |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
9 |
|
10 |
-
|
11 |
-
|
12 |
-
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
13 |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
14 |
|
15 |
-
|
16 |
-
|
17 |
-
|
18 |
-
examples=[
|
19 |
-
os.path.join(os.path.dirname(__file__),
|
20 |
-
"example/video.mp4")],
|
21 |
-
cache_examples=True)
|
22 |
|
23 |
-
|
24 |
-
|
|
|
|
|
25 |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
import gradio as gr
|
2 |
+
import os
|
3 |
+
import cv2
|
4 |
+
import numpy as np
|
5 |
|
6 |
+
import esim_py
|
7 |
+
from infererence import process_events, Ev2Hands
|
8 |
+
from settings import OUTPUT_HEIGHT, OUTPUT_WIDTH, REF_PERIOD
|
9 |
|
10 |
|
11 |
+
os.makedirs("temp", exist_ok=True)
|
12 |
+
ev2hands = Ev2Hands()
|
13 |
+
|
14 |
+
|
15 |
+
def get_frames(video_in, trim_in):
|
16 |
+
cap = cv2.VideoCapture(video_in)
|
17 |
+
|
18 |
+
fps = cap.get(cv2.CAP_PROP_FPS)
|
19 |
+
stop_frame = int(trim_in * fps)
|
20 |
+
|
21 |
+
print("video fps: " + str(fps))
|
22 |
+
|
23 |
+
frames = []
|
24 |
+
i = 0
|
25 |
+
while(cap.isOpened()):
|
26 |
+
ret, frame = cap.read()
|
27 |
+
if not ret:
|
28 |
+
break
|
29 |
+
|
30 |
+
frame = cv2.resize(frame, (OUTPUT_WIDTH, OUTPUT_HEIGHT))
|
31 |
+
frames.append(frame)
|
32 |
+
|
33 |
+
if i > stop_frame:
|
34 |
+
break
|
35 |
+
|
36 |
+
i += 1
|
37 |
+
|
38 |
+
|
39 |
+
cap.release()
|
40 |
+
|
41 |
+
return frames, fps
|
42 |
+
|
43 |
+
|
44 |
+
|
45 |
+
def infer(video_inp, trim_in, threshold):
|
46 |
+
frames, fps = get_frames(video_inp, trim_in)
|
47 |
+
ts_s = 1 / fps
|
48 |
+
ts_ns = ts_s * 1e9 # convert s to ns
|
49 |
+
|
50 |
+
POS_THRESHOLD = NEG_THRESHOLD = threshold
|
51 |
+
|
52 |
+
esim = esim_py.EventSimulator(POS_THRESHOLD, NEG_THRESHOLD, REF_PERIOD, 1e-4, True)
|
53 |
+
is_init = False
|
54 |
+
|
55 |
+
event_frame_vid_path = 'temp/event_video.mp4'
|
56 |
+
prediction_vid_path = 'temp/prediction_video.mp4'
|
57 |
+
|
58 |
+
height, width, _ = frames[0].shape
|
59 |
+
event_video = cv2.VideoWriter(event_frame_vid_path, cv2.VideoWriter_fourcc(*'mp4v'), fps, (width, height))
|
60 |
+
prediction_video = cv2.VideoWriter(prediction_vid_path, cv2.VideoWriter_fourcc(*'mp4v'), fps, (width, height))
|
61 |
+
|
62 |
+
for idx, frame in enumerate(frames):
|
63 |
+
frame_gray = cv2.cvtColor(frame, cv2.COLOR_BGR2GRAY)
|
64 |
+
frame_log = np.log(frame_gray.astype("float32") / 255 + 1e-4)
|
65 |
+
|
66 |
+
current_ts_ns = idx * ts_ns
|
67 |
+
|
68 |
+
if not is_init:
|
69 |
+
esim.init(frame_log, current_ts_ns)
|
70 |
+
is_init = True
|
71 |
+
continue
|
72 |
+
|
73 |
+
events = esim.generateEventFromCVImage(frame_log, current_ts_ns)
|
74 |
+
data = process_events(events)
|
75 |
+
|
76 |
+
prediction_frame = ev2hands(data)
|
77 |
+
event_frame = data['event_frame'].cpu().numpy().astype(dtype=np.uint8)
|
78 |
+
|
79 |
+
|
80 |
+
event_video.write(event_frame)
|
81 |
+
prediction_video.write(prediction_frame)
|
82 |
+
|
83 |
+
event_video.release()
|
84 |
+
prediction_video.release()
|
85 |
+
|
86 |
+
return event_frame_vid_path, prediction_vid_path
|
87 |
+
|
88 |
|
89 |
+
title = """
|
90 |
+
<div style="text-align: center; max-width: 700px; margin: 0 auto;">
|
91 |
+
<div
|
92 |
+
style="
|
93 |
+
display: inline-flex;
|
94 |
+
align-items: center;
|
95 |
+
gap: 0.8rem;
|
96 |
+
font-size: 1.75rem;
|
97 |
+
"
|
98 |
+
>
|
99 |
+
<h1 style="font-weight: 900; margin-bottom: 7px; margin-top: 5px;">
|
100 |
+
Pix2Pix Video
|
101 |
+
</h1>
|
102 |
+
</div>
|
103 |
+
<p style="margin-bottom: 10px; font-size: 94%">
|
104 |
+
Apply Instruct Pix2Pix Diffusion to a video
|
105 |
+
</p>
|
106 |
+
</div>
|
107 |
+
"""
|
108 |
|
109 |
+
article = """
|
110 |
+
|
111 |
+
<div class="footer">
|
112 |
+
<p>
|
113 |
+
Examples by <a href="https://twitter.com/CitizenPlain" target="_blank">Nathan Shipley</a> •
|
114 |
+
Follow <a href="https://twitter.com/fffiloni" target="_blank">Sylvain Filoni</a> for future updates 🤗
|
115 |
+
</p>
|
116 |
+
</div>
|
117 |
+
<div id="may-like-container" style="display: flex;justify-content: center;flex-direction: column;align-items: center;margin-bottom: 30px;">
|
118 |
+
<p>You may also like: </p>
|
119 |
+
<div id="may-like-content" style="display:flex;flex-wrap: wrap;align-items:center;height:20px;">
|
120 |
+
|
121 |
+
<svg height="20" width="162" style="margin-left:4px;margin-bottom: 6px;">
|
122 |
+
<a href="https://huggingface.co/spaces/timbrooks/instruct-pix2pix" target="_blank">
|
123 |
+
<image href="https://img.shields.io/badge/🤗 Spaces-Instruct_Pix2Pix-blue" src="https://img.shields.io/badge/🤗 Spaces-Instruct_Pix2Pix-blue.png" height="20"/>
|
124 |
+
</a>
|
125 |
+
</svg>
|
126 |
+
|
127 |
+
</div>
|
128 |
+
|
129 |
+
</div>
|
130 |
+
|
131 |
+
"""
|
132 |
|
133 |
+
with gr.Blocks(css='style.css') as demo:
|
134 |
+
with gr.Column(elem_id="col-container"):
|
135 |
+
gr.HTML(title)
|
136 |
+
with gr.Row():
|
137 |
+
with gr.Column():
|
138 |
+
video_inp = gr.Video(label="Video source", elem_id="input-vid")
|
139 |
+
with gr.Row():
|
140 |
+
trim_in = gr.Slider(label="Cut video at (s)", minimum=1, maximum=5, step=1, value=1)
|
141 |
+
threshold = gr.Slider(label="Event Threshold", minimum=0.1, maximum=1, step=0.05, value=0.5)
|
142 |
|
143 |
+
with gr.Column():
|
144 |
+
event_frame_out = gr.Video(label="Event Frame", elem_id="video-output")
|
145 |
+
prediction_out = gr.Video(label="Ev2Hands result", elem_id="video-output")
|
|
|
|
|
|
|
|
|
146 |
|
147 |
+
gr.HTML("""
|
148 |
+
<a style="display:inline-block" href="https://huggingface.co/spaces/fffiloni/Pix2Pix-Video?duplicate=true"><img src="https://img.shields.io/badge/-Duplicate%20Space-blue?labelColor=white&style=flat&logo=data:image/png;base64,iVBORw0KGgoAAAANSUhEUgAAABAAAAAQCAYAAAAf8/9hAAAAAXNSR0IArs4c6QAAAP5JREFUOE+lk7FqAkEURY+ltunEgFXS2sZGIbXfEPdLlnxJyDdYB62sbbUKpLbVNhyYFzbrrA74YJlh9r079973psed0cvUD4A+4HoCjsA85X0Dfn/RBLBgBDxnQPfAEJgBY+A9gALA4tcbamSzS4xq4FOQAJgCDwV2CPKV8tZAJcAjMMkUe1vX+U+SMhfAJEHasQIWmXNN3abzDwHUrgcRGmYcgKe0bxrblHEB4E/pndMazNpSZGcsZdBlYJcEL9Afo75molJyM2FxmPgmgPqlWNLGfwZGG6UiyEvLzHYDmoPkDDiNm9JR9uboiONcBXrpY1qmgs21x1QwyZcpvxt9NS09PlsPAAAAAElFTkSuQmCC&logoWidth=14" alt="Duplicate Space"></a>
|
149 |
+
work with longer videos / skip the queue:
|
150 |
+
""", elem_id="duplicate-container")
|
151 |
|
152 |
+
submit_btn = gr.Button("Run Ev2Hands")
|
153 |
+
|
154 |
+
inputs = [video_inp, trim_in, threshold]
|
155 |
+
outputs = [event_frame_out, prediction_out]
|
156 |
+
gr.HTML(article)
|
157 |
+
|
158 |
+
submit_btn.click(infer, inputs, outputs)
|
159 |
+
|
160 |
+
demo.queue(max_size=12).launch(server_name="0.0.0.0", server_port=7860)
|
demo.py → infererence.py
RENAMED
@@ -138,84 +138,52 @@ def demo(net, device, data):
|
|
138 |
return frames
|
139 |
|
140 |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
141 |
|
142 |
-
def main():
|
143 |
-
arg_parser.demo()
|
144 |
-
os.makedirs('outputs', exist_ok=True)
|
145 |
-
|
146 |
-
device = torch.device('cpu')
|
147 |
-
|
148 |
-
net = TEHNetWrapper(device=device)
|
149 |
-
|
150 |
-
save_path = os.environ['CHECKPOINT_PATH']
|
151 |
-
batch_size = int(os.environ['BATCH_SIZE'])
|
152 |
-
|
153 |
-
checkpoint = torch.load(save_path, map_location=device)
|
154 |
-
net.load_state_dict(checkpoint['state_dict'], strict=True)
|
155 |
-
|
156 |
-
renderer = pyrender.OffscreenRenderer(viewport_width=OUTPUT_WIDTH, viewport_height=OUTPUT_HEIGHT)
|
157 |
-
|
158 |
-
scene = pyrender.Scene(ambient_light=(0.3, 0.3, 0.3))
|
159 |
-
light = pyrender.DirectionalLight(color=[1.0, 1.0, 1.0], intensity=0.8)
|
160 |
-
light_pose = np.eye(4)
|
161 |
-
light_pose[:3, 3] = np.array([0, -1, 1])
|
162 |
-
scene.add(light, pose=light_pose)
|
163 |
-
light_pose[:3, 3] = np.array([0, 1, 1])
|
164 |
-
scene.add(light, pose=light_pose)
|
165 |
-
light_pose[:3, 3] = np.array([1, 1, 2])
|
166 |
-
scene.add(light, pose=light_pose)
|
167 |
-
|
168 |
-
rot = trimesh.transformations.rotation_matrix(np.radians(180), [1, 0, 0])
|
169 |
-
|
170 |
-
mano_hands = net.hands
|
171 |
-
|
172 |
-
# camera = cv2.VideoCapture(0)
|
173 |
-
input_video_stream = cv2.VideoCapture('video.mp4')
|
174 |
-
|
175 |
-
|
176 |
-
|
177 |
-
video_fps = 25
|
178 |
-
video = cv2.VideoWriter('outputs/video.mp4', cv2.VideoWriter_fourcc(*'mp4v'), video_fps, (3 * OUTPUT_WIDTH, OUTPUT_HEIGHT))
|
179 |
-
|
180 |
-
|
181 |
-
POS_THRESHOLD = 0.5
|
182 |
-
NEG_THRESHOLD = 0.5
|
183 |
-
REF_PERIOD = 0.000
|
184 |
-
|
185 |
-
esim = esim_py.EventSimulator(POS_THRESHOLD, NEG_THRESHOLD, REF_PERIOD, 1e-4, True)
|
186 |
-
|
187 |
-
|
188 |
-
fps = cv2.CAP_PROP_FPS
|
189 |
-
ts_s = 1 / fps
|
190 |
-
ts_ns = ts_s * 1e9 # convert s to ns
|
191 |
-
|
192 |
-
is_init = False
|
193 |
-
idx = 0
|
194 |
-
while True:
|
195 |
-
_, frame_bgr = input_video_stream.read()
|
196 |
-
frame_bgr = cv2.resize(frame_bgr, (OUTPUT_WIDTH, OUTPUT_HEIGHT))
|
197 |
-
frame_gray = cv2.cvtColor(frame_bgr, cv2.COLOR_BGR2GRAY)
|
198 |
-
frame_log = np.log(frame_gray.astype("float32") / 255 + 1e-4)
|
199 |
-
height, width = frame_log.shape[:2]
|
200 |
-
|
201 |
-
current_ts_ns = idx * ts_ns
|
202 |
-
|
203 |
-
if not is_init:
|
204 |
-
esim.init(frame_log, current_ts_ns)
|
205 |
-
is_init = True
|
206 |
-
idx += 1
|
207 |
-
|
208 |
-
continue
|
209 |
-
idx += 1
|
210 |
-
|
211 |
-
events = esim.generateEventFromCVImage(frame_log, current_ts_ns)
|
212 |
-
data = process_events(events)
|
213 |
-
|
214 |
-
event_frame = data['event_frame'].cpu().numpy().astype(dtype=np.uint8)
|
215 |
-
|
216 |
-
cv2.imwrite(f"outputs/event_frame_{idx}.png", event_frame)
|
217 |
|
218 |
-
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
219 |
|
220 |
frame = demo(net=net, device=device, data=data)[0]
|
221 |
seg_mask = frame['seg_mask']
|
@@ -231,33 +199,14 @@ def main():
|
|
231 |
pred_meshes = trimesh.util.concatenate(pred_meshes)
|
232 |
pred_meshes.apply_transform(rot)
|
233 |
|
234 |
-
camera = MAIN_CAMERA
|
235 |
-
|
236 |
-
nc = pyrender.Node(camera=camera, matrix=np.eye(4))
|
237 |
-
scene.add_node(nc)
|
238 |
|
239 |
mesh_node = pyrender.Node(mesh=pyrender.Mesh.from_trimesh(pred_meshes))
|
240 |
scene.add_node(mesh_node)
|
241 |
pred_rgb, depth = renderer.render(scene)
|
242 |
scene.remove_node(mesh_node)
|
243 |
-
scene.remove_node(nc)
|
244 |
|
245 |
pred_rgb = cv2.cvtColor(pred_rgb, cv2.COLOR_RGB2BGR)
|
246 |
pred_rgb[pred_rgb == 255] = 0
|
247 |
|
248 |
-
|
249 |
-
video.write(img_stack)
|
250 |
-
|
251 |
-
cv2.imshow('image', img_stack)
|
252 |
-
c = cv2.waitKey(1)
|
253 |
-
|
254 |
-
if c == ord('q'):
|
255 |
-
video.release()
|
256 |
-
exit(0)
|
257 |
-
|
258 |
-
video.release()
|
259 |
-
|
260 |
-
|
261 |
-
if __name__ == '__main__':
|
262 |
-
main()
|
263 |
|
|
|
138 |
return frames
|
139 |
|
140 |
|
141 |
+
class Ev2Hands:
|
142 |
+
def __init__(self) -> None:
|
143 |
+
arg_parser.demo()
|
144 |
+
device = torch.device('cpu')
|
145 |
+
net = TEHNetWrapper(device=device)
|
146 |
+
|
147 |
+
save_path = os.environ['CHECKPOINT_PATH']
|
148 |
+
|
149 |
+
checkpoint = torch.load(save_path, map_location=device)
|
150 |
+
net.load_state_dict(checkpoint['state_dict'], strict=True)
|
151 |
+
|
152 |
+
renderer = pyrender.OffscreenRenderer(viewport_width=OUTPUT_WIDTH, viewport_height=OUTPUT_HEIGHT)
|
153 |
+
|
154 |
+
scene = pyrender.Scene(ambient_light=(0.3, 0.3, 0.3))
|
155 |
+
light = pyrender.DirectionalLight(color=[1.0, 1.0, 1.0], intensity=0.8)
|
156 |
+
light_pose = np.eye(4)
|
157 |
+
light_pose[:3, 3] = np.array([0, -1, 1])
|
158 |
+
scene.add(light, pose=light_pose)
|
159 |
+
light_pose[:3, 3] = np.array([0, 1, 1])
|
160 |
+
scene.add(light, pose=light_pose)
|
161 |
+
light_pose[:3, 3] = np.array([1, 1, 2])
|
162 |
+
scene.add(light, pose=light_pose)
|
163 |
+
|
164 |
+
camera = MAIN_CAMERA
|
165 |
+
nc = pyrender.Node(camera=camera, matrix=np.eye(4))
|
166 |
+
scene.add_node(nc)
|
167 |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
168 |
|
169 |
+
rot = trimesh.transformations.rotation_matrix(np.radians(180), [1, 0, 0])
|
170 |
+
|
171 |
+
mano_hands = net.hands
|
172 |
+
|
173 |
+
self.net = net
|
174 |
+
self.device = device
|
175 |
+
self.mano_hands = mano_hands
|
176 |
+
self.rot = rot
|
177 |
+
self.renderer = renderer
|
178 |
+
self.scene = scene
|
179 |
+
|
180 |
+
def __call__(self, data):
|
181 |
+
net = self.net
|
182 |
+
device = self.device
|
183 |
+
mano_hands = self.mano_hands
|
184 |
+
rot = self.rot
|
185 |
+
renderer = self.renderer
|
186 |
+
scene = self.scene
|
187 |
|
188 |
frame = demo(net=net, device=device, data=data)[0]
|
189 |
seg_mask = frame['seg_mask']
|
|
|
199 |
pred_meshes = trimesh.util.concatenate(pred_meshes)
|
200 |
pred_meshes.apply_transform(rot)
|
201 |
|
|
|
|
|
|
|
|
|
202 |
|
203 |
mesh_node = pyrender.Node(mesh=pyrender.Mesh.from_trimesh(pred_meshes))
|
204 |
scene.add_node(mesh_node)
|
205 |
pred_rgb, depth = renderer.render(scene)
|
206 |
scene.remove_node(mesh_node)
|
|
|
207 |
|
208 |
pred_rgb = cv2.cvtColor(pred_rgb, cv2.COLOR_RGB2BGR)
|
209 |
pred_rgb[pred_rgb == 255] = 0
|
210 |
|
211 |
+
return pred_rgb
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
212 |
|