Update app.py

app.py

@@ -40,7 +40,6 @@ CLIP_SIZE = 224
 BATCH_SIZE = 1
 SEED = 42
 
-
 # Set the device (GPU or CPU)
 device = torch.device("cuda" if torch.cuda.is_available() else "cpu")
 # pretrained Model
@@ -49,6 +48,14 @@ MODEL_TRANSFORMER = 'google/vivit-b-16x2'
 model_path = 'vivit_pytorch_loss051.pt'
 data_path = 'signs'
 
+# Custom CSS to control output video size
+custom_css = """
+#landmarked_video {
+    max-height: 350px;
+    width: auto;
+}
+"""
+
 # Create Mediapipe Objects
 mp_drawing = mp.solutions.drawing_utils
 mp_drawing_styles = mp.solutions.drawing_styles
@@ -304,7 +311,7 @@ def translate_sign_language(gesture):
             
     return gesture_translation , prod_video_path # frame_generator
 
-with gr.Blocks() as demo:
+with gr.Blocks(css=custom_css) as demo:
     gr.Markdown("# Indian Sign Language Translation App")
     
     # Gesture recognition Tab
@@ -318,8 +325,8 @@ with gr.Blocks() as demo:
                 # Display the landmarked video
                 video_output = gr.Video(interactive=False, autoplay=True, 
                                         streaming=False, label="Landmarked Gesture",
-                                        elem_id="landmarked_video",
-                                        style={"max-height": "350px", "width": "auto"})
+                                        elem_id="landmarked_video"
+                                        )
             
         with gr.Row(variant="panel"): # equal_height=False, show_progress=True
             with gr.Column(scale=1, variant="panel"):