-
Notifications
You must be signed in to change notification settings - Fork 1
/
main.py
211 lines (191 loc) · 8.2 KB
/
main.py
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
207
208
209
210
211
#!/usr/bin/env python
# _*_ codign:utf8 _*_
"""====================================
@Author:Sadam·Sadik
@Email：[email protected]
@Date：2024/2/29
@Software: PyCharm
@disc:
======================================="""
import os
import cv2
import gradio as gr
import numpy as np
from moviepy.editor import VideoClip
from PIL import Image
from core.hash import get_combined_hash
# global variable definition:
webui_server_url = 'http://218.31.113.195:57860'
out_dir = 'api_out'
out_dir_i2i = os.path.join(out_dir, 'img2img')
videoWidth = 1920
videoHeight = 1080
videoFPS = 30
def get_fp(frame_img_fp, params):
hash_value = get_combined_hash(frame_img_fp, params)
save_fp = os.path.join(out_dir_i2i, f'{hash_value}.png')
return save_fp
def frame_to_video_clip(frame_path):
# 读取图片
frame = np.asarray(Image.open(frame_path))
# 将图片转换为 VideoClip 对象
clip = VideoClip(frame, duration=1 / 25)
return clip
def frames_to_video(frame_paths, output_path, fps):
# 创建一个 VideoClip 对象，将所有 frame 连接起来
clip = VideoClip([frame_to_video_clip(frame_path) for frame_path in frame_paths], concat=True)
# 设置视频的帧率
clip = clip.set_fps(fps)
# 保存视频为 MP4 文件
clip.write_videofile(output_path)
def generate_video(frames: list[slice]):
fourcc = cv2.VideoWriter_fourcc(*"mp4v")
"""
参数2 VideoWriter_fourcc为视频编解码器
fourcc意为四字符代码（Four-Character Codes），顾名思义，该编码由四个字符组成,下面是VideoWriter_fourcc对象一些常用的参数,注意：字符顺序不能弄混
cv2.VideoWriter_fourcc('I', '4', '2', '0'),该参数是YUV编码类型，文件名后缀为.avi
cv2.VideoWriter_fourcc('P', 'I', 'M', 'I'),该参数是MPEG-1编码类型，文件名后缀为.avi
cv2.VideoWriter_fourcc('X', 'V', 'I', 'D'),该参数是MPEG-4编码类型，文件名后缀为.avi
cv2.VideoWriter_fourcc('T', 'H', 'E', 'O'),该参数是Ogg Vorbis,文件名后缀为.ogv
cv2.VideoWriter_fourcc('F', 'L', 'V', '1'),该参数是Flash视频，文件名后缀为.flv
cv2.VideoWriter_fourcc('m', 'p', '4', 'v') 文件名后缀为.mp4
"""
generated_video_fp = "generated_video.mp4"
print(f"Generate params:[fps:{videoFPS}, width:{videoWidth}, height:{videoHeight}, fourcc:{fourcc}]", )
video = cv2.VideoWriter(generated_video_fp, fourcc, int(videoFPS),
(int(videoWidth), int(videoHeight))) # 创建视频流对象-格式一
# frame_paths = []
for i, frame in enumerate(frames):
frame_fp, flag = frame
# print(i, frame_fp, flag)
image = cv2.imread(frame_fp)
video.write(image)
print(i, flag)
# frames_to_video(frame_paths, generated_video_fp, videoFPS)
return generated_video_fp
def on_frame_selected(evt: gr.SelectData, frames: list[slice], seed: int, prompt: str):
print(f"OnFrameSelected:[{evt.index} from {evt.target}, Seed:{seed}]")
print(evt.value)
"""
{
'image': {
'path': '/private/var/folders/by/gmwsqd_x01q02gssqxb8ry140000gn/T/gradio/b5b7c1932bd4383da5e747ad11f5c39399222c33/image.png',
'url': 'http://127.0.0.1:7860/file=/private/var/folders/by/gmwsqd_x01q02gssqxb8ry140000gn/T/gradio/b5b7c1932bd4383da5e747ad11f5c39399222c33/image.png',
'size': None,
'orig_name': None,
'mime_type': None,
'is_stream': False
},
'caption': 'frame33'
}
"""
res = []
for i, frame in enumerate(frames):
frame_fp, flag = frame
# print(i, frame_fp, flag)
if i == evt.index:
res.append(["api_out/img2img/2ffc5feecb71e7ada52357c330d0b1a7d6b677fbdf2fe6fe71789dadb6366d7c.png", flag])
else:
res.append([frame_fp, flag])
#
file_name = "saved_frame.jpg"
# 保存图像
# cv2.imwrite(file_name, org_frame)
params = {
"prompt": prompt,
"seed": seed,
"steps": 20,
"width": videoWidth,
"height": videoHeight,
"denoising_strength": 0.5,
"n_iter": 1,
}
# save_fp = get_fp(file_name, params)
# if os.path.exists(save_fp):
# # 更新GUI上的画面
# new_frameElem.update(filename=save_fp, size=(1200, 400))
# else:
# q.put([file_name, params])
# generated_fp = frame_diffusion(file_name, params)
# print("已经处理好了:", file_name, params, generated_fp)
return res
def on_input_video_change(input_video_fp):
global videoWidth, videoHeight, videoFPS
# /private/var/folders/by/gmwsqd_x01q02gssqxb8ry140000gn/T/gradio/51b4179c3e0c8063abb954c87fe139a3a1bec014/2521442.jpeg.mp4
print("on_input_video_change:", input_video_fp)
cap = cv2.VideoCapture(input_video_fp)
videoWidth = cap.get(3)
videoHeight = cap.get(4)
videoFPS = cap.get(5)
videoTotalFrames = int(cap.get(7))
print("video info: [fps:{},total_frame:{}]".format(videoFPS, videoTotalFrames))
print("frameSlider.maximum:", seedSlider.maximum)
frames = []
for frame_index in range(videoTotalFrames):
cap.set(cv2.CAP_PROP_POS_FRAMES, frame_index)
ret, org_frame = cap.read()
print("On Input Video processing: frame{}, {:.2f}%".format(frame_index, frame_index / videoTotalFrames * 100))
if ret:
# 转换为 RGB 颜色格式
rgb_frame = cv2.cvtColor(org_frame, cv2.COLOR_BGR2RGB)
# 将 OpenCV 帧转换为 numpy.ndarray
frame_ndarray = np.asarray(rgb_frame)
frames.append([frame_ndarray, "frame{}".format(frame_index)])
# 释放Video
cap.release()
info = f'''
## Video Info
* Frame: {videoWidth}x{videoHeight}
* Fps: {videoFPS} Frame Count:{videoTotalFrames} Duration: {videoTotalFrames / videoFPS}s
'''
return frames, info
# print("video frames num:", input_v.get_num_frames())
with gr.Blocks(js="extensions/js-extension.js") as demo:
gr.State()
# with gr.Row():
# with gr.Column(scale=1):
#
# with gr.Column(scale=3):
gr.Markdown(
'''
# Frame Diffusion
Start typing below and then click **Run** to see the output.
'''
)
with gr.Row():
with gr.Column(scale=1):
inputVideoView = gr.Video(label="InputVideo", height=480)
videoInfoView = gr.Markdown(
'''
## Video Info
'''
)
promptElem = gr.Text(label="prompt", interactive=True, lines=3, value="1girl, blue hair")
seedSlider = gr.Slider(label="Seed", elem_id="frameSlider", randomize=True, minimum=1, maximum=3290305185)
progress = gr.Progress()
btn = gr.Button("Start Generation", interactive=True)
with gr.Column(scale=3):
with gr.Tab("Generated Frames"):
with gr.Row():
originalFrameGallery = gr.Gallery(label="Original Frames", columns=6, height=400, interactive=False)
with gr.Row():
generatedFrameGallery = gr.Gallery(label="Generated Frames", columns=6, height=400,
interactive=False)
with gr.Tab("Generated Video"):
out = gr.Video(label="Out", show_download_button=True, height=760)
# with gr.Row():
# with gr.Column(scale=1):
# frameCountView = gr.Number(label="Frame Count", interactive=False, elem_id="frameCount")
# gr.Markdown("Video Info:\nwidth:200px\nheight:300px")
# gr.Image(value="api_out/img2img/img2img-20240229-172125-0.png", interactive=True)
# # with gr.Column(scale=2):
# with gr.Row():
# gr.Gallery(
# value=["api_out/img2img/img2img-20240229-172054-0.png", "api_out/img2img/img2img-20240229-171824-0.png"])
# 开始绑定事件
inputVideoView.change(on_input_video_change, inputs=[inputVideoView], outputs=[originalFrameGallery, videoInfoView])
originalFrameGallery.select(on_frame_selected, inputs=[originalFrameGallery, seedSlider, promptElem],
outputs=[generatedFrameGallery])
btn.click(fn=generate_video, inputs=[generatedFrameGallery], outputs=out)
if __name__ == '__main__':
demo.launch()