Update app.py

app.py

@@ -613,8 +613,10 @@ with gr.Blocks() as demo:
     gr.Markdown(f"<div style='text-align:center;'>{text}</div>")
 
     with gr.Column():
+
+        gr.Markdown("Step1: Upload a Source Video", elem_id="ref_title")
+        
         video_input = gr.Video(label="Upload Video", elem_id="my-video1")
-        get_info_btn = gr.Button("Extract First Frame", elem_id="my-btn")
         
         gr.Examples(
             examples=[
@@ -630,6 +632,9 @@ with gr.Blocks() as demo:
             elem_id="my-btn2"
         )
 
+        gr.Markdown("Step2: Extract the First Frame & Click for Segmentation", elem_id="ref_title")
+        get_info_btn = gr.Button("Extract First Frame", elem_id="my-btn")
+
         image_output = gr.Image(
             label="First Frame Segmentation",
             interactive=True,
@@ -642,6 +647,7 @@ with gr.Blocks() as demo:
             )
             clear_btn = gr.Button("Clear All Clicks")
 
+        gr.Markdown("Step3: Track to Get Video Mask", elem_id="ref_title")
         with gr.Row(elem_id="my-btn"):
             n_frames_slider = gr.Slider(
                 minimum=1, maximum=201, value=81, step=1, label="Tracking Frames (4N+1)"
@@ -649,7 +655,7 @@ with gr.Blocks() as demo:
             track_btn = gr.Button("Tracking")
         video_output = gr.Video(label="Tracking Result", elem_id="my-video")
 
-        gr.Markdown("Reference Image & Mask (SAM2 Points)", elem_id="ref_title")
+        gr.Markdown("Step4: Upload a Reference Image", elem_id="ref_title")
 
         ref_image_input = gr.Image(
             label="Upload Reference Image", elem_id="ref-image", interactive=True
@@ -670,6 +676,8 @@ with gr.Blocks() as demo:
             label="You can upload or choose a reference image below to retexture.",
             elem_id="my-btn3"
         )
+
+        gr.Markdown("Step5: Click for Segmentation", elem_id="ref_title")
         ref_image_display = gr.Image(
             label="Reference Mask Segmentation",
             elem_id="ref-mask",