Update app.py

app.py

@@ -1,4 +1,4 @@
-# app.py (Complete Final Version - Fixed resize_masks SyntaxError)
+# app.py (Complete Final Version - Fixed SyntaxError in overlap loop)
 
 import gradio as gr
 import torch
@@ -167,6 +167,7 @@ def classify_image_clip(image_pil):
         traceback.print_exc()
         return "Error during CLIP processing", {"Error": 1.0}
 
+# --- CORRECTED process_car_image Function ---
 def process_car_image(image_np_bgr, damage_threshold, part_threshold):
     """
     Runs damage and part segmentation (YOLOv8), calculates overlap, visualizes.
@@ -215,38 +216,21 @@ def process_car_image(image_np_bgr, damage_threshold, part_threshold):
         yolo_end_time = time.time()
         logger.info(f" YOLO predictions took {yolo_end_time - yolo_start_time:.2f}s")
 
-        # --- 3. Resize Masks (Corrected Function Definition) ---
+        # --- 3. Resize Masks ---
         def resize_masks(masks_tensor, target_h, target_w):
             """Resizes masks tensor to target H, W using CPU numpy and OpenCV."""
-            masks_np_bool = masks_tensor.cpu().numpy().astype(bool) # Move to CPU *before* resizing
-
-            # Handle empty tensor or already correct size
-            if masks_np_bool.shape[0] == 0:
-                return np.array([]) # Return empty numpy array
-            if masks_np_bool.ndim == 3 and masks_np_bool.shape[1] == target_h and masks_np_bool.shape[2] == target_w:
-                return masks_np_bool # Return if already correct size
-
-            # Ensure masks_np_bool is 3D [N, H, W] even if only one mask
-            if masks_np_bool.ndim == 2: # Handle case of single mask output [H, W]
-                 masks_np_bool = np.expand_dims(masks_np_bool, axis=0)
-                 logger.warning("Detected 2D mask input, expanding to 3D for resize loop.")
-
-            # Check dimensions *before* logging resize message
-            if masks_np_bool.ndim != 3:
-                 logger.error(f"Unexpected mask dimension: {masks_np_bool.ndim}. Expected 3D [N, H, W]. Cannot resize.")
-                 return np.array([]) # Return empty if shape is wrong
-
-            # Proceed with resizing if necessary
-            # logger.info(f"Resizing {masks_np_bool.shape[0]} masks from {masks_np_bool.shape[1:]} to {(target_h, target_w)}") # Optional verbose log
+            masks_np_bool = masks_tensor.cpu().numpy().astype(bool)
+            if masks_np_bool.shape[0] == 0: return np.array([])
+            if masks_np_bool.ndim == 3 and masks_np_bool.shape[1] == target_h and masks_np_bool.shape[2] == target_w: return masks_np_bool
+            if masks_np_bool.ndim == 2: masks_np_bool = np.expand_dims(masks_np_bool, axis=0)
+            if masks_np_bool.ndim != 3: logger.error(f"Unexpected mask dim: {masks_np_bool.ndim}"); return np.array([])
             resized_masks_list = []
             for i in range(masks_np_bool.shape[0]):
-                mask = masks_np_bool[i] # Get the single [H, W] mask
-                # Resize needs uint8
+                mask = masks_np_bool[i]
                 mask_resized = cv2.resize(mask.astype(np.uint8), (target_w, target_h), interpolation=cv2.INTER_NEAREST)
-                resized_masks_list.append(mask_resized.astype(bool)) # Append boolean mask
-            return np.array(resized_masks_list) # Return numpy array [N, target_h, target_w]
+                resized_masks_list.append(mask_resized.astype(bool))
+            return np.array(resized_masks_list)
 
-        # --- Perform resizing ---
         resize_start_time = time.time()
         damage_masks_np = resize_masks(damage_masks_raw, img_h, img_w)
         part_masks_np = resize_masks(part_masks_raw, img_h, img_w)
@@ -259,10 +243,61 @@ def process_car_image(image_np_bgr, damage_threshold, part_threshold):
         overlap_start_time = time.time()
         if damage_masks_np.shape[0] > 0 and part_masks_np.shape[0] > 0:
             overlap_threshold = 0.4
-            for i in range(len(damage_masks_np)):
-                 damage_mask = damage_masks_np[i]; damage_class_id = damage_classes_ids_cpu[i]; try: damage_name = DAMAGE_CLASSES[damage_class_id]; except IndexError: logger.warning(f"Invalid damage ID {damage_class_id} during overlap"); continue; damage_area = np.sum(damage_mask); if damage_area < 10: continue; max_overlap = 0; assigned_part_name = "Unknown / Outside Parts";
-                 for j in range(len(part_masks_np)): part_mask = part_masks_np[j]; part_class_id = part_classes_ids_cpu[j]; try: part_name = CAR_PART_CLASSES[part_class_id]; except IndexError: logger.warning(f"Invalid part ID {part_class_id} during overlap"); continue; intersection = np.logical_and(damage_mask, part_mask); overlap_ratio = np.sum(intersection) / damage_area if damage_area > 0 else 0; if overlap_ratio > max_overlap: max_overlap = overlap_ratio; if max_overlap >= overlap_threshold: assigned_part_name = part_name;
-                 assignment_desc = f"{damage_name} in {assigned_part_name}"; if assigned_part_name == "Unknown / Outside Parts": assignment_desc += f" (Overlap < {overlap_threshold*100:.0f}%)"; final_assignments.append(assignment_desc)
+            # *** CORRECTED OVERLAP LOOP SYNTAX ***
+            for i in range(len(damage_masks_np)): # Iterate through each detected damage
+                damage_mask = damage_masks_np[i]
+                damage_class_id = damage_classes_ids_cpu[i]
+
+                # Try getting damage name, skip if ID is invalid
+                try:
+                    damage_name = DAMAGE_CLASSES[damage_class_id]
+                except IndexError:
+                    logger.warning(f"Invalid damage ID {damage_class_id} found during overlap check. Skipping this damage.")
+                    continue # Go to the next damage mask
+
+                # Check damage area (only if name was valid)
+                damage_area = np.sum(damage_mask)
+                if damage_area < 10: # Skip tiny masks
+                    continue
+
+                # Initialize for finding best overlapping part for this damage
+                max_overlap = 0
+                assigned_part_name = "Unknown / Outside Parts" # Default
+
+                # Inner loop for parts
+                for j in range(len(part_masks_np)):
+                    part_mask = part_masks_np[j]
+                    part_class_id = part_classes_ids_cpu[j]
+                    try:
+                        part_name = CAR_PART_CLASSES[part_class_id]
+                    except IndexError:
+                        logger.warning(f"Invalid part ID {part_class_id} during overlap check.")
+                        continue # Skip this part
+
+                    intersection = np.logical_and(damage_mask, part_mask)
+                    overlap_ratio = np.sum(intersection) / damage_area if damage_area > 0 else 0
+
+                    if overlap_ratio > max_overlap:
+                        max_overlap = overlap_ratio
+                        # Assign based on threshold condition within the inner loop
+                        if max_overlap >= overlap_threshold:
+                            assigned_part_name = part_name
+                        else:
+                            # If max overlap is < threshold even for best part, keep default
+                            assigned_part_name = "Unknown / Outside Parts"
+
+
+                # Append assignment result after checking all parts for the current damage
+                assignment_desc = f"{damage_name} in {assigned_part_name}"
+                if assigned_part_name == "Unknown / Outside Parts" and max_overlap > 0: # Add detail if there was *some* overlap
+                    assignment_desc += f" (Max Overlap < {overlap_threshold*100:.0f}%)"
+                elif assigned_part_name == "Unknown / Outside Parts":
+                     assignment_desc += " (No overlap)" # Clarify if zero overlap
+
+                final_assignments.append(assignment_desc)
+            # *** END OF CORRECTED OVERLAP LOOP ***
+
+        # Handle cases with no damages or no parts
         elif damage_masks_np.shape[0] > 0: final_assignments.append(f"{len(damage_masks_np)} damages found, but no parts detected/matched above threshold {part_threshold}.")
         elif part_masks_np.shape[0] > 0: final_assignments.append(f"No damages detected above threshold {damage_threshold}.")
         else: final_assignments.append(f"No damages or parts detected above thresholds.")