Update app.py

app.py

@@ -2,40 +2,67 @@ import torch
 import numpy as np
 import gradio as gr
 from PIL import Image
+import time
 
 # Device configuration
 device = torch.device('cuda' if torch.cuda.is_available() else 'cpu')
 
-# Load YOLOv5s model (smallest and fastest variant)
+# Load optimized YOLOv5s model
 model = torch.hub.load('ultralytics/yolov5', 'yolov5s', pretrained=True).to(device)
 
-# Enable half-precision for CUDA devices
+# Performance optimizations
+model.conf = 0.5  # Confidence threshold (adjust for speed/accuracy balance)
 if device.type == 'cuda':
-    model.half()
+    model.half()  # FP16 precision
 
-def detect_objects(image):
-    # Convert numpy array to PIL Image
-    image_pil = Image.fromarray(image)
+def process_frame(image):
+    """Process single frame with error handling"""
+    if image is None:
+        return None
     
-    # Perform inference without gradient calculation
-    with torch.no_grad():
-        results = model(image_pil)
+    try:
+        # Convert numpy array to PIL Image
+        image_pil = Image.fromarray(image)
+        
+        # Perform inference
+        with torch.no_grad():
+            results = model(image_pil)
+        
+        # Render results
+        rendered_images = results.render()
+        return np.array(rendered_images[0]) if rendered_images else image
     
-    # Render detections using optimized YOLOv5 method
-    rendered_images = results.render()
-    
-    # Return the first rendered image as numpy array
-    return np.array(rendered_images[0]) if rendered_images else image
+    except Exception as e:
+        print(f"Processing error: {e}")
+        return image
 
-# Gradio interface
-iface = gr.Interface(
-    fn=detect_objects,
-    inputs=gr.Image(type="numpy", label="Upload Image"),
-    outputs=gr.Image(type="numpy", label="Detected Objects"),
-    title="High-Speed Object Detection with YOLOv5s",
-    description="Optimized for speed using YOLOv5s and GPU acceleration.",
-    allow_flagging="never",
-    examples=["spring_street_after.jpg", "pexels-hikaique-109919.jpg"]
-)
+with gr.Blocks(title="Real-Time Object Detection") as app:
+    gr.Markdown("# 🚀 Real-Time Object Detection with Dual Input")
+    gr.Markdown("Supports live webcam streaming and image uploads")
+    
+    with gr.Tabs():
+        with gr.TabItem("📷 Live Camera"):
+            with gr.Row():
+                webcam_input = gr.Webcam(label="Live Feed", streaming=True)
+                live_output = gr.Image(label="Processed Feed", streaming=True)
+            webcam_input.change(process_frame, webcam_input, live_output)
+            
+        with gr.TabItem("🖼️ Image Upload"):
+            with gr.Row():
+                upload_input = gr.Image(type="numpy", label="Upload Image")
+                upload_output = gr.Image(label="Detection Result")
+            upload_input.change(process_frame, upload_input, upload_output)
+            
+    gr.Markdown("### ⚙️ Performance Settings")
+    with gr.Accordion("Advanced Settings", open=False):
+        gr.Slider(minimum=0.1, maximum=0.9, value=0.5, 
+                 label="Confidence Threshold", interactive=True)
+        gr.Checkbox(label="Enable FP16 Acceleration", value=True)
 
-iface.launch()
+# Configure queue and launch
+app.queue(concurrency_count=4, max_size=20).launch(
+    server_name="0.0.0.0",
+    server_port=7860,
+    share=False,
+    enable_queue=True
+)