Update app.py

app.py

@@ -227,19 +227,18 @@ def click_to_similarity_in_same_image(
     return marked_ref, heatmap_pil, overlay_pil, overlay_boxes_pil
 
 # ----------------------------
-# Gradio UI (+ Start button, + Model dropdown)
+# Gradio UI (Manual-only processing)
 # ----------------------------
 with gr.Blocks(theme=gr.themes.Soft(), title="DINOv3 Single-Image Patch Similarity") as demo:
     gr.Markdown("# 🦖 DINOv3 Single-Image Patch Similarity")
-    gr.Markdown("## Running on CPU-only Space, feature extraction after uploading an image can take a moment")
-    gr.Markdown("Upload one image, then **click anywhere** to highlight the most similar regions in the *same* image.")
+    gr.Markdown("Upload one image, adjust settings, then press **▶️ Start processing**. Click on the processed image to find similar regions.")
 
     app_state = gr.State()
 
     with gr.Row():
         with gr.Column(scale=1):
             input_image = gr.Image(
-                label="Image (click anywhere)",
+                label="Image (click anywhere after processing)",
                 type="pil",
                 value="https://images.squarespace-cdn.com/content/v1/607f89e638219e13eee71b1e/1684821560422-SD5V37BAG28BURTLIXUQ/michael-sum-LEpfefQf4rU-unsplash.jpg"
             )
@@ -254,14 +253,14 @@ with gr.Blocks(theme=gr.themes.Soft(), title="DINOv3 Single-Image Patch Similari
                     ["viridis", "magma", "plasma", "inferno", "turbo", "cividis"],
                     value="viridis", label="Colormap",
                 )
-            # NEW: Backbone selector (default = smaller/faster ViT-S/16+)
+            # Backbone selector (default = smaller/faster ViT-S/16+)
             model_choice = gr.Dropdown(
                 choices=AVAILABLE_MODELS,
                 value=DEFAULT_MODEL_ID,
                 label="Backbone (DINOv3)",
                 info="ViT-S/16+ is smaller & faster; ViT-H/16+ is larger.",
             )
-            # Start processing button
+            # Start processing button (manual trigger)
             with gr.Row():
                 start_btn = gr.Button("▶️ Start processing", variant="primary")
 
@@ -271,7 +270,7 @@ with gr.Blocks(theme=gr.themes.Soft(), title="DINOv3 Single-Image Patch Similari
             box_radius = gr.Slider(0, 10, value=1, step=1, label="Box radius (patches)")
 
     with gr.Row():
-        marked_image = gr.Image(label="Click marker", interactive=False)
+        marked_image = gr.Image(label="Click marker / Preview", interactive=False)
         heatmap_output = gr.Image(label="Similarity heatmap", interactive=False)
     with gr.Row():
         overlay_output = gr.Image(label="Overlay (image ⊕ heatmap)", interactive=False)
@@ -284,7 +283,8 @@ with gr.Blocks(theme=gr.themes.Soft(), title="DINOv3 Single-Image Patch Similari
             model = get_model(model_id)
             _current_model_id = model_id
 
-    def _on_upload_or_slider_change(img: Image.Image, long_side: int, model_id: str, progress=gr.Progress(track_tqdm=True)):
+    # Manual feature extraction (only runs on Start button)
+    def _run_extraction(img: Image.Image, long_side: int, model_id: str, progress=gr.Progress(track_tqdm=True)):
         if img is None:
             return None, None
         _ensure_model(model_id)
@@ -293,6 +293,7 @@ with gr.Blocks(theme=gr.themes.Soft(), title="DINOv3 Single-Image Patch Similari
         progress(1, desc="Done!")
         return st["img"], st
 
+    # Clicking on processed image to compute similarities
     def _on_click(st, a: float, m: str, excl: int, k: int, box_rad: int, evt: gr.SelectData):
         if not st or evt is None:
             return None, None, None, None
@@ -302,20 +303,29 @@ with gr.Blocks(theme=gr.themes.Soft(), title="DINOv3 Single-Image Patch Similari
             box_radius_patches=int(box_rad),
         )
 
-    # Wire events
-    inputs_for_update = [input_image, target_long_side, model_choice]
-    outputs_for_update = [marked_image, app_state]
-
-    # Auto triggers (kept)
-    input_image.upload(_on_upload_or_slider_change, inputs=inputs_for_update, outputs=outputs_for_update)
-    target_long_side.change(_on_upload_or_slider_change, inputs=inputs_for_update, outputs=outputs_for_update)
-    model_choice.change(_on_upload_or_slider_change, inputs=inputs_for_update, outputs=outputs_for_update)
-    demo.load(_on_upload_or_slider_change, inputs=inputs_for_update, outputs=outputs_for_update)  # Process default image on load
+    # On image change: just preview and clear outputs/state (NO extraction)
+    def _on_image_changed(img: Image.Image):
+        if img is None:
+            return None, None, None, None, None
+        return img, None, None, None, None
+
+    # ---------- Wiring (Manual mode) ----------
+    # Do NOT auto-run on upload/slider/model change or on app load.
+    # Only the Start button triggers extraction.
+    start_btn.click(
+        _run_extraction,
+        inputs=[input_image, target_long_side, model_choice],
+        outputs=[marked_image, app_state],
+    )
 
-    # Manual trigger via button (kept)
-    start_btn.click(_on_upload_or_slider_change, inputs=inputs_for_update, outputs=outputs_for_update)
+    # When a new image is picked, show it as preview and clear old results.
+    input_image.change(
+        _on_image_changed,
+        inputs=[input_image],
+        outputs=[marked_image, app_state, heatmap_output, overlay_output, overlay_boxes_output],
+    )
 
-    # Click to compute similarities
+    # Keep click handler the same.
     marked_image.select(
         _on_click,
         inputs=[app_state, alpha, cmap, exclude_r, topk, box_radius],