Upload app.py

app.py

@@ -517,7 +517,7 @@ def process_image(image, source_language, target_language, mode, font, font_colo
     # Detect text and get results
     img_with_boxes, img_bubbles, img_removed_text, merged_results, mask = detect_and_show_text(reader, image)
 
-    if mode == "Basic":
+    if mode == "Basic (speech bubbles only)":
         img_inpainted = img_removed_text
     else:
         img_inpainted = process(image, mask, num_inference_steps=num_inference_steps)
@@ -563,11 +563,11 @@ def update_translations(image, edited_texts, translations_list, img_removed_text
     return np.array(final_result)
 
 
-with gr.Blocks(title="Nativ - Demo") as demo:
+with gr.Blocks(title="Nativ Demo - Localize text within Comics") as demo:
     # Store translations list in state
     translations_state = gr.State([])
     
-    gr.Markdown("# Nativ - Demo")
+    gr.Markdown("# Nativ Demo - Localize text within Comics")
     
     with gr.Row():
         with gr.Column():
@@ -585,8 +585,8 @@ with gr.Blocks(title="Nativ - Demo") as demo:
             )
             # Toggle for mode selection
             localization_mode = gr.Radio(
-                choices=["Basic", "Advanced"],
-                value="Basic",
+                choices=["Basic (speech bubbles only)", "Advanced (all text)"],
+                value="Basic (speech bubbles only)",
                 label="Localization Mode"
             )
             font_selector_i = gr.Dropdown(
@@ -654,10 +654,10 @@ with gr.Blocks(title="Nativ - Demo") as demo:
     # Add examples here
     gr.Examples(
         examples=[
-            ["assets/chinese.png", "Simplified Chinese", "English", "Basic", "omniscript_bold", "#000000"],
-            ["assets/chinese.png", "Simplified Chinese", "English", "Advanced", "Ldfcomicsansbold", "#d31515"],
-            ["assets/korean.png", "Korean", "Spanish", "Basic", "omniscript_bold", "#000000"],
-            ["assets/chinese.png", "English", "French", "Basic", "omniscript_bold", "#000000"],
+            ["assets/chinese.png", "Simplified Chinese", "English", "Basic (speech bubbles only)", "omniscript_bold", "#000000"],
+            ["assets/chinese.png", "Simplified Chinese", "English", "Advanced (all text)", "Ldfcomicsansbold", "#d31515"],
+            ["assets/korean.png", "Korean", "Spanish", "Basic (speech bubbles only)", "omniscript_bold", "#000000"],
+            ["assets/chinese.png", "English", "French", "Basic (speech bubbles only)", "omniscript_bold", "#000000"],
         ],
         inputs=[input_image, source_language, target_language, localization_mode, font_selector_i, font_color_picker_i]
     )