remove mask

app.py

@@ -98,6 +98,7 @@ net = build_model(hypar, device)
 def inference(file_paths, logs):
     """
     Process up to 3 images uploaded via the file uploader.
+    Only the image with background removed is returned.
     """
     start_time = time.time()
     logs = logs or ""
@@ -107,7 +108,7 @@ def inference(file_paths, logs):
 
     # Limit to a maximum of 3 images
     image_paths = file_paths[:3]
-    processed_pairs = []
+    processed_images = []
     for path in image_paths:
         image_tensor, orig_size = load_image(path, hypar)
         mask = predict(net, image_tensor, orig_size, hypar, device)
@@ -115,12 +116,11 @@ def inference(file_paths, logs):
         im_rgb = Image.open(path).convert("RGB")
         im_rgba = im_rgb.copy()
         im_rgba.putalpha(pil_mask)
-        processed_pairs.append([im_rgba, pil_mask])
+        processed_images.append(im_rgba)
 
     elapsed = round(time.time() - start_time, 2)
-    final_images = [img for pair in processed_pairs for img in pair]
-    logs += f"Processed {len(processed_pairs)} image(s) in {elapsed} second(s).\n"
-    return final_images, logs, logs
+    logs += f"Processed {len(processed_images)} image(s) in {elapsed} second(s).\n"
+    return processed_images, logs, logs
 
 title = "Highly Accurate Dichotomous Image Segmentation"
 description = (
@@ -142,7 +142,7 @@ interface = gr.Interface(
         gr.State()
     ],
     outputs=[
-        gr.Gallery(label="Output (rgba + mask)"),
+        gr.Gallery(label="Output (Background Removed)"),
         gr.State(),
         gr.Textbox(label="Logs", lines=6)
     ],