final commit

gradio_app.py

@@ -41,6 +41,8 @@ download_models()
 def run_epic_inference(video_path, caption, motion_type):
     temp_input_path = "/app/temp_input.mp4"
     output_dir = f"/app/output_{motion_type}"
+    
+    video_output_path = f"{output_dir}/masked_videos/temp_input.mp4"
     traj_name = motion_type
     traj_txt = f"/app/inference/v2v_data/test/trajs/{traj_name}.txt"
 
@@ -83,8 +85,8 @@ def run_epic_inference(video_path, caption, motion_type):
 
     # Locate the output video
     output_video = Path(output_dir) / f"amalfi-coast_traj_{traj_name}.mp4"
-    if output_video.exists():
-        return logs, str(output_video)
+    if video_output_path.exists():
+        return logs, str(video_output_path)
     else:
         return f"Inference succeeded but no output video found in {output_dir}", None
 

inference/v2v_data/demo.py

@@ -130,6 +130,7 @@ class GetAnchorVideos:
             window_size=opts.window_size,
             overlap=opts.overlap,
         ).to(opts.device)
+        
         frames = (
             torch.from_numpy(frames).permute(0, 3, 1, 2).to(opts.device) * 2.0 - 1.0
         )  # 49 576 1024 3 -> 49 3 576 1024, [-1,1]