重新排版了下。。。

app.py

@@ -10,7 +10,7 @@ import skvideo.io
 
 # 全局设置
 DEFAULT_CLASSES = ["background", "aeroplane", "bicycle", "bird", "boat", "bottle", "bus", "car", "cat", "chair",
-                                   "cow", "diningtable", "dog", "horse", "motorbike", "person", "pottedplant", "sheep", "sofa", "train", "tvmonitor"]
+                   "cow", "diningtable", "dog", "horse", "motorbike", "person", "pottedplant", "sheep", "sofa", "train", "tvmonitor"]
 COLORS = np.random.uniform(0, 255, size=(len(DEFAULT_CLASSES), 3))
 DISPLAY_INTERVAL = 30
 CONFIDENCE_LEVEL = 0.5
@@ -61,9 +61,10 @@ def process_video(input_video_path, selected_classes):
                 originEndY = int(endY*origin_h/h)
 
                 if DEFAULT_CLASSES[idx] in selected_classes:
-                    label = "{}: {:.2f}%".format(DEFAULT_CLASSES[idx], confidence * 100)
+                    label = "{}: {:.2f}%".format(
+                        DEFAULT_CLASSES[idx], confidence * 100)
                     cv2.rectangle(origin_frame, (originStartX, originStartY),
-                                (originEndX, originEndY), COLORS[idx], 2)
+                                  (originEndX, originEndY), COLORS[idx], 2)
 
                     y = originStartY - 15 if originStartY - 15 > 15 else originStartY + 15
                     cv2.putText(origin_frame, label, (originStartX, y),
@@ -132,18 +133,29 @@ def main():
     # 定义Gradio界面
     with gr.Blocks() as demo:
         with gr.Row():
-            input_video = gr.Video(label="Video Input")
-            processed_frames = gr.Image(label="Live Preview")
-            output_video = gr.Video(label="Video Output")
-
+            gr.Markdown(
+                """
+    # Hello! Welcome to the Object-Detection Demo.
+    """)
         with gr.Row():
+            input_video = gr.Video(label="Video Input")
             with gr.Column():
+                processed_frames = gr.Image(label="Live Preview")
+                gr.Markdown(
+                    """
+    Note: Output video will be visible after previewing the video.
+    """)
+                output_video = gr.Video(label="Video Output")
+
+        with gr.Column():
+            with gr.Row():
                 input_classes = gr.Dropdown(
                     DEFAULT_CLASSES, value=DEFAULT_CLASSES, multiselect=True, label="Detection Classes", info="Select the classes that you want to detect."
                 )
                 examples = gr.Examples(
                     [f"test/{file.name}" for file in os.scandir('./test')], inputs=input_video)
-            process_video_btn = gr.Button("process video")
+            process_video_btn = gr.Button(
+                "process video", style="width: 100px; height: 30px;")
 
         process_video_btn.click(fn=process_video, inputs=[
                                 input_video, input_classes], outputs=[processed_frames, output_video])