Muhammad Taqi Raza
commited on
Commit
Β·
767f3bb
1
Parent(s):
e7707d9
reverting gradio
Browse files- gradio_app.py +215 -215
gradio_app.py
CHANGED
@@ -1,174 +1,3 @@
|
|
1 |
-
# import os
|
2 |
-
# import subprocess
|
3 |
-
# from datetime import datetime
|
4 |
-
# from pathlib import Path
|
5 |
-
# import gradio as gr
|
6 |
-
|
7 |
-
# # -----------------------------
|
8 |
-
# # Setup paths and env
|
9 |
-
# # -----------------------------
|
10 |
-
# HF_HOME = "/app/hf_cache"
|
11 |
-
# os.environ["HF_HOME"] = HF_HOME
|
12 |
-
# os.environ["TRANSFORMERS_CACHE"] = HF_HOME
|
13 |
-
# os.makedirs(HF_HOME, exist_ok=True)
|
14 |
-
|
15 |
-
# PRETRAINED_DIR = "/app/pretrained"
|
16 |
-
# os.makedirs(PRETRAINED_DIR, exist_ok=True)
|
17 |
-
|
18 |
-
|
19 |
-
# # -----------------------------
|
20 |
-
# # Step 1: Optional Model Download
|
21 |
-
# # -----------------------------
|
22 |
-
# def download_models():
|
23 |
-
# expected_model = os.path.join(PRETRAINED_DIR, "RAFT/raft-things.pth")
|
24 |
-
# if not Path(expected_model).exists():
|
25 |
-
# print("βοΈ Downloading pretrained models...")
|
26 |
-
# try:
|
27 |
-
# subprocess.check_call(["bash", "download/download_models.sh"])
|
28 |
-
# print("β
Models downloaded.")
|
29 |
-
# except subprocess.CalledProcessError as e:
|
30 |
-
# print(f"β Model download failed: {e}")
|
31 |
-
# else:
|
32 |
-
# print("β
Pretrained models already exist.")
|
33 |
-
|
34 |
-
|
35 |
-
# download_models()
|
36 |
-
|
37 |
-
|
38 |
-
# # -----------------------------
|
39 |
-
# # Step 2: Inference Logic
|
40 |
-
# # -----------------------------
|
41 |
-
|
42 |
-
# def run_epic_inference(video_path, caption, motion_type):
|
43 |
-
# temp_input_path = "/app/temp_input.mp4"
|
44 |
-
# output_dir = f"/app/output_anchor"
|
45 |
-
# video_output_path = f"{output_dir}/masked_videos/output.mp4"
|
46 |
-
# traj_name = motion_type
|
47 |
-
# traj_txt = f"/app/inference/v2v_data/test/trajs/{traj_name}.txt"
|
48 |
-
# # Save uploaded video
|
49 |
-
# if video_path:
|
50 |
-
# os.system(f"cp '{video_path}' {temp_input_path}")
|
51 |
-
|
52 |
-
# command = [
|
53 |
-
# "python", "/app/inference/v2v_data/inference.py",
|
54 |
-
# "--video_path", temp_input_path,
|
55 |
-
# "--stride", "1",
|
56 |
-
# "--out_dir", output_dir,
|
57 |
-
# "--radius_scale", "1",
|
58 |
-
# "--camera", "target",
|
59 |
-
# "--mask",
|
60 |
-
# "--target_pose", "0", "30", "-0.6", "0", "0",
|
61 |
-
# "--traj_txt", traj_txt,
|
62 |
-
# "--save_name", "output",
|
63 |
-
# "--mode", "gradual",
|
64 |
-
# ]
|
65 |
-
|
66 |
-
# # Run inference command
|
67 |
-
# try:
|
68 |
-
# result = subprocess.run(command, capture_output=True, text=True, check=True)
|
69 |
-
# print("Getting Anchor Videos run successfully.")
|
70 |
-
# logs = result.stdout
|
71 |
-
# except subprocess.CalledProcessError as e:
|
72 |
-
# logs = f"β Inference failed:\n{e.stderr}"
|
73 |
-
# return logs, None
|
74 |
-
|
75 |
-
# # Locate the output video
|
76 |
-
# if video_output_path:
|
77 |
-
# return logs, str(video_output_path)
|
78 |
-
# else:
|
79 |
-
# return f"Inference succeeded but no output video found in {output_dir}", None
|
80 |
-
# def print_output_directory(out_dir):
|
81 |
-
# result = ""
|
82 |
-
# for root, dirs, files in os.walk(out_dir):
|
83 |
-
# level = root.replace(out_dir, '').count(os.sep)
|
84 |
-
# indent = ' ' * 4 * level
|
85 |
-
# result += f"{indent}{os.path.basename(root)}/"
|
86 |
-
# sub_indent = ' ' * 4 * (level + 1)
|
87 |
-
# for f in files:
|
88 |
-
# result += f"{sub_indent}{f}\n"
|
89 |
-
# return result
|
90 |
-
|
91 |
-
# def inference(video_path, caption, motion_type):
|
92 |
-
# logs, video_masked = run_epic_inference(video_path, caption, motion_type)
|
93 |
-
|
94 |
-
# MODEL_PATH="/app/pretrained/CogVideoX-5b-I2V"
|
95 |
-
|
96 |
-
# ckpt_steps=500
|
97 |
-
# ckpt_dir="/app/out/EPiC_pretrained"
|
98 |
-
# ckpt_file=f"checkpoint-{ckpt_steps}.pt"
|
99 |
-
# ckpt_path=f"{ckpt_dir}/{ckpt_file}"
|
100 |
-
|
101 |
-
# video_root_dir= f"/app/output_anchor"
|
102 |
-
# out_dir=f"/app/output"
|
103 |
-
|
104 |
-
# command = [
|
105 |
-
# "python", "/app/inference/cli_demo_camera_i2v_pcd.py",
|
106 |
-
# "--video_root_dir", video_root_dir,
|
107 |
-
# "--base_model_path", MODEL_PATH,
|
108 |
-
# "--controlnet_model_path", ckpt_path,
|
109 |
-
# "--output_path", out_dir,
|
110 |
-
# "--start_camera_idx", "0",
|
111 |
-
# "--end_camera_idx", "8",
|
112 |
-
# "--controlnet_weights", "1.0",
|
113 |
-
# "--controlnet_guidance_start", "0.0",
|
114 |
-
# "--controlnet_guidance_end", "0.4",
|
115 |
-
# "--controlnet_input_channels", "3",
|
116 |
-
# "--controlnet_transformer_num_attn_heads", "4",
|
117 |
-
# "--controlnet_transformer_attention_head_dim", "64",
|
118 |
-
# "--controlnet_transformer_out_proj_dim_factor", "64",
|
119 |
-
# "--controlnet_transformer_out_proj_dim_zero_init",
|
120 |
-
# "--vae_channels", "16",
|
121 |
-
# "--num_frames", "49",
|
122 |
-
# "--controlnet_transformer_num_layers", "8",
|
123 |
-
# "--infer_with_mask",
|
124 |
-
# "--pool_style", "max",
|
125 |
-
# "--seed", "43"
|
126 |
-
# ]
|
127 |
-
|
128 |
-
# # Run the command
|
129 |
-
# result = subprocess.run(command, capture_output=True, text=True)
|
130 |
-
# if result.returncode == 0:
|
131 |
-
# print("Inference completed successfully.")
|
132 |
-
# else:
|
133 |
-
# print(f"Error occurred during inference: {result.stderr}")
|
134 |
-
|
135 |
-
# # Print output directory contents
|
136 |
-
# logs = result.stdout
|
137 |
-
# result = print_output_directory(out_dir)
|
138 |
-
|
139 |
-
# return logs+result, str(f"{out_dir}/00000_43_out.mp4")
|
140 |
-
|
141 |
-
# # output 43
|
142 |
-
# # output/ 00000_43_out.mp4
|
143 |
-
# # 00000_43_reference.mp4
|
144 |
-
# # 00000_43_out_reference.mp4
|
145 |
-
|
146 |
-
# # -----------------------------
|
147 |
-
# # Step 3: Create Gradio UI
|
148 |
-
# # -----------------------------
|
149 |
-
# demo = gr.Interface(
|
150 |
-
# fn=inference,
|
151 |
-
# inputs=[
|
152 |
-
# gr.Video(label="Upload Video (MP4)"),
|
153 |
-
# gr.Textbox(label="Caption", placeholder="e.g., Amalfi coast with boats"),
|
154 |
-
# gr.Dropdown(
|
155 |
-
# choices=["zoom_in", "rotate", "orbit", "pan", "loop1"],
|
156 |
-
# label="Camera Motion Type",
|
157 |
-
# value="zoom_in",
|
158 |
-
# ),
|
159 |
-
# ],
|
160 |
-
# outputs=[gr.Textbox(label="Inference Logs"), gr.Video(label="Generated Video")],
|
161 |
-
# title="π¬ EPiC: Efficient Video Camera Control",
|
162 |
-
# description="Upload a video, describe the scene, and apply cinematic camera motion using pretrained EPiC models.",
|
163 |
-
# )
|
164 |
-
|
165 |
-
# # -----------------------------
|
166 |
-
# # Step 4: Launch App
|
167 |
-
# # -----------------------------
|
168 |
-
# if __name__ == "__main__":
|
169 |
-
# demo.launch(server_name="0.0.0.0", server_port=7860)
|
170 |
-
|
171 |
-
|
172 |
import os
|
173 |
import subprocess
|
174 |
from datetime import datetime
|
@@ -186,6 +15,7 @@ os.makedirs(HF_HOME, exist_ok=True)
|
|
186 |
PRETRAINED_DIR = "/app/pretrained"
|
187 |
os.makedirs(PRETRAINED_DIR, exist_ok=True)
|
188 |
|
|
|
189 |
# -----------------------------
|
190 |
# Step 1: Optional Model Download
|
191 |
# -----------------------------
|
@@ -201,69 +31,75 @@ def download_models():
|
|
201 |
else:
|
202 |
print("β
Pretrained models already exist.")
|
203 |
|
|
|
204 |
download_models()
|
205 |
|
|
|
206 |
# -----------------------------
|
207 |
# Step 2: Inference Logic
|
208 |
# -----------------------------
|
209 |
-
|
|
|
210 |
temp_input_path = "/app/temp_input.mp4"
|
211 |
-
output_dir = "/app/output_anchor"
|
212 |
video_output_path = f"{output_dir}/masked_videos/output.mp4"
|
213 |
-
|
|
|
214 |
# Save uploaded video
|
215 |
if video_path:
|
216 |
os.system(f"cp '{video_path}' {temp_input_path}")
|
217 |
|
218 |
-
try:
|
219 |
-
theta, phi, r, x, y = target_pose.strip().split()
|
220 |
-
except ValueError:
|
221 |
-
return f"β Invalid target pose format. Use: ΞΈ Ο r x y", None
|
222 |
-
|
223 |
command = [
|
224 |
-
|
225 |
-
|
226 |
-
|
227 |
-
|
228 |
-
|
229 |
-
|
230 |
-
|
231 |
-
|
232 |
-
|
233 |
-
|
|
|
234 |
]
|
235 |
|
|
|
236 |
try:
|
237 |
result = subprocess.run(command, capture_output=True, text=True, check=True)
|
|
|
238 |
logs = result.stdout
|
239 |
except subprocess.CalledProcessError as e:
|
240 |
logs = f"β Inference failed:\n{e.stderr}"
|
241 |
return logs, None
|
242 |
|
243 |
-
|
244 |
-
|
|
|
|
|
|
|
245 |
def print_output_directory(out_dir):
|
246 |
result = ""
|
247 |
for root, dirs, files in os.walk(out_dir):
|
248 |
level = root.replace(out_dir, '').count(os.sep)
|
249 |
indent = ' ' * 4 * level
|
250 |
-
result += f"{indent}{os.path.basename(root)}
|
251 |
sub_indent = ' ' * 4 * (level + 1)
|
252 |
for f in files:
|
253 |
result += f"{sub_indent}{f}\n"
|
254 |
return result
|
255 |
|
256 |
-
def inference(video_path,
|
257 |
-
logs, video_masked = run_epic_inference(video_path,
|
258 |
|
259 |
-
MODEL_PATH
|
260 |
-
ckpt_steps = 500
|
261 |
-
ckpt_dir = "/app/out/EPiC_pretrained"
|
262 |
-
ckpt_file = f"checkpoint-{ckpt_steps}.pt"
|
263 |
-
ckpt_path = f"{ckpt_dir}/{ckpt_file}"
|
264 |
|
265 |
-
|
266 |
-
|
|
|
|
|
|
|
|
|
|
|
267 |
|
268 |
command = [
|
269 |
"python", "/app/inference/cli_demo_camera_i2v_pcd.py",
|
@@ -282,19 +118,30 @@ def inference(video_path, num_frames, fps, target_pose, mode):
|
|
282 |
"--controlnet_transformer_out_proj_dim_factor", "64",
|
283 |
"--controlnet_transformer_out_proj_dim_zero_init",
|
284 |
"--vae_channels", "16",
|
285 |
-
"--num_frames",
|
286 |
"--controlnet_transformer_num_layers", "8",
|
287 |
"--infer_with_mask",
|
288 |
"--pool_style", "max",
|
289 |
-
"--seed", "43"
|
290 |
-
"--fps", str(fps)
|
291 |
]
|
292 |
|
|
|
293 |
result = subprocess.run(command, capture_output=True, text=True)
|
294 |
-
|
295 |
-
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
296 |
|
297 |
-
|
|
|
|
|
|
|
298 |
|
299 |
# -----------------------------
|
300 |
# Step 3: Create Gradio UI
|
@@ -303,15 +150,14 @@ demo = gr.Interface(
|
|
303 |
fn=inference,
|
304 |
inputs=[
|
305 |
gr.Video(label="Upload Video (MP4)"),
|
306 |
-
gr.
|
307 |
-
gr.
|
308 |
-
|
309 |
-
|
310 |
-
|
311 |
-
|
312 |
-
gr.Textbox(label="Inference Logs"),
|
313 |
-
gr.Video(label="Generated Video")
|
314 |
],
|
|
|
315 |
title="π¬ EPiC: Efficient Video Camera Control",
|
316 |
description="Upload a video, describe the scene, and apply cinematic camera motion using pretrained EPiC models.",
|
317 |
)
|
@@ -321,3 +167,157 @@ demo = gr.Interface(
|
|
321 |
# -----------------------------
|
322 |
if __name__ == "__main__":
|
323 |
demo.launch(server_name="0.0.0.0", server_port=7860)
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
import os
|
2 |
import subprocess
|
3 |
from datetime import datetime
|
|
|
15 |
PRETRAINED_DIR = "/app/pretrained"
|
16 |
os.makedirs(PRETRAINED_DIR, exist_ok=True)
|
17 |
|
18 |
+
|
19 |
# -----------------------------
|
20 |
# Step 1: Optional Model Download
|
21 |
# -----------------------------
|
|
|
31 |
else:
|
32 |
print("β
Pretrained models already exist.")
|
33 |
|
34 |
+
|
35 |
download_models()
|
36 |
|
37 |
+
|
38 |
# -----------------------------
|
39 |
# Step 2: Inference Logic
|
40 |
# -----------------------------
|
41 |
+
|
42 |
+
def run_epic_inference(video_path, caption, motion_type):
|
43 |
temp_input_path = "/app/temp_input.mp4"
|
44 |
+
output_dir = f"/app/output_anchor"
|
45 |
video_output_path = f"{output_dir}/masked_videos/output.mp4"
|
46 |
+
traj_name = motion_type
|
47 |
+
traj_txt = f"/app/inference/v2v_data/test/trajs/{traj_name}.txt"
|
48 |
# Save uploaded video
|
49 |
if video_path:
|
50 |
os.system(f"cp '{video_path}' {temp_input_path}")
|
51 |
|
|
|
|
|
|
|
|
|
|
|
52 |
command = [
|
53 |
+
"python", "/app/inference/v2v_data/inference.py",
|
54 |
+
"--video_path", temp_input_path,
|
55 |
+
"--stride", "1",
|
56 |
+
"--out_dir", output_dir,
|
57 |
+
"--radius_scale", "1",
|
58 |
+
"--camera", "target",
|
59 |
+
"--mask",
|
60 |
+
"--target_pose", "0", "30", "-0.6", "0", "0",
|
61 |
+
"--traj_txt", traj_txt,
|
62 |
+
"--save_name", "output",
|
63 |
+
"--mode", "gradual",
|
64 |
]
|
65 |
|
66 |
+
# Run inference command
|
67 |
try:
|
68 |
result = subprocess.run(command, capture_output=True, text=True, check=True)
|
69 |
+
print("Getting Anchor Videos run successfully.")
|
70 |
logs = result.stdout
|
71 |
except subprocess.CalledProcessError as e:
|
72 |
logs = f"β Inference failed:\n{e.stderr}"
|
73 |
return logs, None
|
74 |
|
75 |
+
# Locate the output video
|
76 |
+
if video_output_path:
|
77 |
+
return logs, str(video_output_path)
|
78 |
+
else:
|
79 |
+
return f"Inference succeeded but no output video found in {output_dir}", None
|
80 |
def print_output_directory(out_dir):
|
81 |
result = ""
|
82 |
for root, dirs, files in os.walk(out_dir):
|
83 |
level = root.replace(out_dir, '').count(os.sep)
|
84 |
indent = ' ' * 4 * level
|
85 |
+
result += f"{indent}{os.path.basename(root)}/"
|
86 |
sub_indent = ' ' * 4 * (level + 1)
|
87 |
for f in files:
|
88 |
result += f"{sub_indent}{f}\n"
|
89 |
return result
|
90 |
|
91 |
+
def inference(video_path, caption, motion_type):
|
92 |
+
logs, video_masked = run_epic_inference(video_path, caption, motion_type)
|
93 |
|
94 |
+
MODEL_PATH="/app/pretrained/CogVideoX-5b-I2V"
|
|
|
|
|
|
|
|
|
95 |
|
96 |
+
ckpt_steps=500
|
97 |
+
ckpt_dir="/app/out/EPiC_pretrained"
|
98 |
+
ckpt_file=f"checkpoint-{ckpt_steps}.pt"
|
99 |
+
ckpt_path=f"{ckpt_dir}/{ckpt_file}"
|
100 |
+
|
101 |
+
video_root_dir= f"/app/output_anchor"
|
102 |
+
out_dir=f"/app/output"
|
103 |
|
104 |
command = [
|
105 |
"python", "/app/inference/cli_demo_camera_i2v_pcd.py",
|
|
|
118 |
"--controlnet_transformer_out_proj_dim_factor", "64",
|
119 |
"--controlnet_transformer_out_proj_dim_zero_init",
|
120 |
"--vae_channels", "16",
|
121 |
+
"--num_frames", "49",
|
122 |
"--controlnet_transformer_num_layers", "8",
|
123 |
"--infer_with_mask",
|
124 |
"--pool_style", "max",
|
125 |
+
"--seed", "43"
|
|
|
126 |
]
|
127 |
|
128 |
+
# Run the command
|
129 |
result = subprocess.run(command, capture_output=True, text=True)
|
130 |
+
if result.returncode == 0:
|
131 |
+
print("Inference completed successfully.")
|
132 |
+
else:
|
133 |
+
print(f"Error occurred during inference: {result.stderr}")
|
134 |
+
|
135 |
+
# Print output directory contents
|
136 |
+
logs = result.stdout
|
137 |
+
result = print_output_directory(out_dir)
|
138 |
+
|
139 |
+
return logs+result, str(f"{out_dir}/00000_43_out.mp4")
|
140 |
|
141 |
+
# output 43
|
142 |
+
# output/ 00000_43_out.mp4
|
143 |
+
# 00000_43_reference.mp4
|
144 |
+
# 00000_43_out_reference.mp4
|
145 |
|
146 |
# -----------------------------
|
147 |
# Step 3: Create Gradio UI
|
|
|
150 |
fn=inference,
|
151 |
inputs=[
|
152 |
gr.Video(label="Upload Video (MP4)"),
|
153 |
+
gr.Textbox(label="Caption", placeholder="e.g., Amalfi coast with boats"),
|
154 |
+
gr.Dropdown(
|
155 |
+
choices=["zoom_in", "rotate", "orbit", "pan", "loop1"],
|
156 |
+
label="Camera Motion Type",
|
157 |
+
value="zoom_in",
|
158 |
+
),
|
|
|
|
|
159 |
],
|
160 |
+
outputs=[gr.Textbox(label="Inference Logs"), gr.Video(label="Generated Video")],
|
161 |
title="π¬ EPiC: Efficient Video Camera Control",
|
162 |
description="Upload a video, describe the scene, and apply cinematic camera motion using pretrained EPiC models.",
|
163 |
)
|
|
|
167 |
# -----------------------------
|
168 |
if __name__ == "__main__":
|
169 |
demo.launch(server_name="0.0.0.0", server_port=7860)
|
170 |
+
|
171 |
+
|
172 |
+
# import os
|
173 |
+
# import subprocess
|
174 |
+
# from datetime import datetime
|
175 |
+
# from pathlib import Path
|
176 |
+
# import gradio as gr
|
177 |
+
|
178 |
+
# # -----------------------------
|
179 |
+
# # Setup paths and env
|
180 |
+
# # -----------------------------
|
181 |
+
# HF_HOME = "/app/hf_cache"
|
182 |
+
# os.environ["HF_HOME"] = HF_HOME
|
183 |
+
# os.environ["TRANSFORMERS_CACHE"] = HF_HOME
|
184 |
+
# os.makedirs(HF_HOME, exist_ok=True)
|
185 |
+
|
186 |
+
# PRETRAINED_DIR = "/app/pretrained"
|
187 |
+
# os.makedirs(PRETRAINED_DIR, exist_ok=True)
|
188 |
+
|
189 |
+
# # -----------------------------
|
190 |
+
# # Step 1: Optional Model Download
|
191 |
+
# # -----------------------------
|
192 |
+
# def download_models():
|
193 |
+
# expected_model = os.path.join(PRETRAINED_DIR, "RAFT/raft-things.pth")
|
194 |
+
# if not Path(expected_model).exists():
|
195 |
+
# print("βοΈ Downloading pretrained models...")
|
196 |
+
# try:
|
197 |
+
# subprocess.check_call(["bash", "download/download_models.sh"])
|
198 |
+
# print("β
Models downloaded.")
|
199 |
+
# except subprocess.CalledProcessError as e:
|
200 |
+
# print(f"β Model download failed: {e}")
|
201 |
+
# else:
|
202 |
+
# print("β
Pretrained models already exist.")
|
203 |
+
|
204 |
+
# download_models()
|
205 |
+
|
206 |
+
# # -----------------------------
|
207 |
+
# # Step 2: Inference Logic
|
208 |
+
# # -----------------------------
|
209 |
+
# def run_epic_inference(video_path, target_pose, mode):
|
210 |
+
# temp_input_path = "/app/temp_input.mp4"
|
211 |
+
# output_dir = "/app/output_anchor"
|
212 |
+
# video_output_path = f"{output_dir}/masked_videos/output.mp4"
|
213 |
+
|
214 |
+
# # Save uploaded video
|
215 |
+
# if video_path:
|
216 |
+
# os.system(f"cp '{video_path}' {temp_input_path}")
|
217 |
+
|
218 |
+
# try:
|
219 |
+
# theta, phi, r, x, y = target_pose.strip().split()
|
220 |
+
# except ValueError:
|
221 |
+
# return f"β Invalid target pose format. Use: ΞΈ Ο r x y", None
|
222 |
+
|
223 |
+
# command = [
|
224 |
+
# "python", "/app/inference/v2v_data/inference.py",
|
225 |
+
# "--video_path", temp_input_path,
|
226 |
+
# "--stride", "1",
|
227 |
+
# "--out_dir", output_dir,
|
228 |
+
# "--radius_scale", "1",
|
229 |
+
# "--camera", "target",
|
230 |
+
# "--mask",
|
231 |
+
# "--target_pose", theta, phi, r, x, y,
|
232 |
+
# "--save_name", "output",
|
233 |
+
# "--mode", mode,
|
234 |
+
# ]
|
235 |
+
|
236 |
+
# try:
|
237 |
+
# result = subprocess.run(command, capture_output=True, text=True, check=True)
|
238 |
+
# logs = result.stdout
|
239 |
+
# except subprocess.CalledProcessError as e:
|
240 |
+
# logs = f"β Inference failed:\n{e.stderr}"
|
241 |
+
# return logs, None
|
242 |
+
|
243 |
+
# return logs, str(video_output_path) if os.path.exists(video_output_path) else (logs, None)
|
244 |
+
|
245 |
+
# def print_output_directory(out_dir):
|
246 |
+
# result = ""
|
247 |
+
# for root, dirs, files in os.walk(out_dir):
|
248 |
+
# level = root.replace(out_dir, '').count(os.sep)
|
249 |
+
# indent = ' ' * 4 * level
|
250 |
+
# result += f"{indent}{os.path.basename(root)}/\n"
|
251 |
+
# sub_indent = ' ' * 4 * (level + 1)
|
252 |
+
# for f in files:
|
253 |
+
# result += f"{sub_indent}{f}\n"
|
254 |
+
# return result
|
255 |
+
|
256 |
+
# def inference(video_path, num_frames, fps, target_pose, mode):
|
257 |
+
# logs, video_masked = run_epic_inference(video_path, target_pose, mode)
|
258 |
+
|
259 |
+
# MODEL_PATH = "/app/pretrained/CogVideoX-5b-I2V"
|
260 |
+
# ckpt_steps = 500
|
261 |
+
# ckpt_dir = "/app/out/EPiC_pretrained"
|
262 |
+
# ckpt_file = f"checkpoint-{ckpt_steps}.pt"
|
263 |
+
# ckpt_path = f"{ckpt_dir}/{ckpt_file}"
|
264 |
+
|
265 |
+
# video_root_dir = "/app/output_anchor"
|
266 |
+
# out_dir = "/app/output"
|
267 |
+
|
268 |
+
# command = [
|
269 |
+
# "python", "/app/inference/cli_demo_camera_i2v_pcd.py",
|
270 |
+
# "--video_root_dir", video_root_dir,
|
271 |
+
# "--base_model_path", MODEL_PATH,
|
272 |
+
# "--controlnet_model_path", ckpt_path,
|
273 |
+
# "--output_path", out_dir,
|
274 |
+
# "--start_camera_idx", "0",
|
275 |
+
# "--end_camera_idx", "8",
|
276 |
+
# "--controlnet_weights", "1.0",
|
277 |
+
# "--controlnet_guidance_start", "0.0",
|
278 |
+
# "--controlnet_guidance_end", "0.4",
|
279 |
+
# "--controlnet_input_channels", "3",
|
280 |
+
# "--controlnet_transformer_num_attn_heads", "4",
|
281 |
+
# "--controlnet_transformer_attention_head_dim", "64",
|
282 |
+
# "--controlnet_transformer_out_proj_dim_factor", "64",
|
283 |
+
# "--controlnet_transformer_out_proj_dim_zero_init",
|
284 |
+
# "--vae_channels", "16",
|
285 |
+
# "--num_frames", str(num_frames),
|
286 |
+
# "--controlnet_transformer_num_layers", "8",
|
287 |
+
# "--infer_with_mask",
|
288 |
+
# "--pool_style", "max",
|
289 |
+
# "--seed", "43",
|
290 |
+
# "--fps", str(fps)
|
291 |
+
# ]
|
292 |
+
|
293 |
+
# result = subprocess.run(command, capture_output=True, text=True)
|
294 |
+
# logs += "\n" + result.stdout
|
295 |
+
# result_dir = print_output_directory(out_dir)
|
296 |
+
|
297 |
+
# return logs + result_dir, str(f"{out_dir}/00000_43_out.mp4")
|
298 |
+
|
299 |
+
# # -----------------------------
|
300 |
+
# # Step 3: Create Gradio UI
|
301 |
+
# # -----------------------------
|
302 |
+
# demo = gr.Interface(
|
303 |
+
# fn=inference,
|
304 |
+
# inputs=[
|
305 |
+
# gr.Video(label="Upload Video (MP4)"),
|
306 |
+
# gr.Slider(minimum=1, maximum=120, value=50, step=1, label="Number of Frames"),
|
307 |
+
# gr.Slider(minimum=1, maximum=90, value=10, step=1, label="FPS"),
|
308 |
+
# gr.Textbox(label="Target Pose (ΞΈ Ο r x y)", placeholder="e.g., 0 30 -0.6 0 0"),
|
309 |
+
# gr.Dropdown(choices=["gradual", "direct", "bullet"], value="gradual", label="Camera Mode"),
|
310 |
+
# ],
|
311 |
+
# outputs=[
|
312 |
+
# gr.Textbox(label="Inference Logs"),
|
313 |
+
# gr.Video(label="Generated Video")
|
314 |
+
# ],
|
315 |
+
# title="π¬ EPiC: Efficient Video Camera Control",
|
316 |
+
# description="Upload a video, describe the scene, and apply cinematic camera motion using pretrained EPiC models.",
|
317 |
+
# )
|
318 |
+
|
319 |
+
# # -----------------------------
|
320 |
+
# # Step 4: Launch App
|
321 |
+
# # -----------------------------
|
322 |
+
# if __name__ == "__main__":
|
323 |
+
# demo.launch(server_name="0.0.0.0", server_port=7860)
|