update

src/audio2vid.py

@@ -5,7 +5,7 @@ from pathlib import Path
 import numpy as np
 import cv2
 import torch
-# import spaces
+import spaces
 from scipy.spatial.transform import Rotation as R
 from scipy.interpolate import interp1d
 
@@ -91,7 +91,7 @@ def get_headpose_temp(input_video):
     
     return pose_arr_smooth
 
-# @spaces.GPU
+@spaces.GPU
 def audio2video(input_audio, ref_img, headpose_video=None, size=512, steps=25, length=150, seed=42):
     fps = 30
     cfg = 3.5

src/vid2vid.py

@@ -7,7 +7,7 @@ from pathlib import Path
 import numpy as np
 import cv2
 import torch
-# import spaces
+import spaces
 from diffusers import AutoencoderKL, DDIMScheduler
 from einops import repeat
 from omegaconf import OmegaConf
@@ -27,21 +27,7 @@ from src.utils.pose_util import project_points_with_trans,  matrix_to_euler_and_
 from src.audio2vid import smooth_pose_seq
 
 
-def parse_args():
-    parser = argparse.ArgumentParser()
-    parser.add_argument("--config", type=str, default='./configs/prompts/animation_facereenac.yaml')
-    parser.add_argument("-W", type=int, default=512)
-    parser.add_argument("-H", type=int, default=512)
-    parser.add_argument("-L", type=int)
-    parser.add_argument("--seed", type=int, default=42)
-    parser.add_argument("--cfg", type=float, default=3.5)
-    parser.add_argument("--steps", type=int, default=25)
-    parser.add_argument("--fps", type=int)
-    args = parser.parse_args()
-
-    return args
-
-# @spaces.GPU
+@spaces.GPU
 def video2video(ref_img, source_video, size=512, steps=25, length=150, seed=42):
     cfg = 3.5