Update app.py

app.py

@@ -1,36 +1,33 @@
 import gradio as gr
 from ultralytics import YOLO
-import tempfile
 import cv2
+import tempfile
+import numpy as np
 
 # Load YOLOv8 model once
-model = YOLO("best.pt")
+model = YOLO("best.pt")  
 
-# Inference on image
+# 📸 Image prediction
 def predict_image(image):
-    results = model.predict(image)
+    results = model.predict(image, imgsz=320, conf=0.4, verbose=False)
     return results[0].plot()
 
-# Inference on video
+# 🎥 Video file prediction
 def predict_video(video_path):
-    # OpenCV video capture
     cap = cv2.VideoCapture(video_path)
     fps = cap.get(cv2.CAP_PROP_FPS)
-    width = int(cap.get(cv2.CAP_PROP_FRAME_WIDTH)) // 2  # Resize for performance
-    height = int(cap.get(cv2.CAP_PROP_FRAME_HEIGHT)) // 2
+    width = int(cap.get(cv2.CAP_PROP_FRAME_WIDTH) // 2)
+    height = int(cap.get(cv2.CAP_PROP_FRAME_HEIGHT) // 2)
 
-    # Output video
     temp_output = tempfile.NamedTemporaryFile(suffix=".mp4", delete=False)
-    fourcc = cv2.VideoWriter_fourcc(*'mp4v')
-    out = cv2.VideoWriter(temp_output.name, fourcc, fps, (width, height))
+    out = cv2.VideoWriter(temp_output.name, cv2.VideoWriter_fourcc(*'mp4v'), fps, (width, height))
 
     while True:
         ret, frame = cap.read()
         if not ret:
             break
-
-        frame = cv2.resize(frame, (width, height))  # Resize to speed up inference
-        results = model.predict(frame, imgsz=480, conf=0.5, verbose=False)  # Lower imgsz = faster
+        frame = cv2.resize(frame, (width, height))
+        results = model.predict(frame, imgsz=320, conf=0.4, verbose=False)
         annotated = results[0].plot()
         out.write(annotated)
 
@@ -38,23 +35,39 @@ def predict_video(video_path):
     out.release()
     return temp_output.name
 
-# Gradio Interface
+# 🔴 Live webcam stream detection
+def live_feed():
+    cap = cv2.VideoCapture(0)
+    while True:
+        ret, frame = cap.read()
+        if not ret:
+            break
+        frame = cv2.resize(frame, (640, 480))
+        results = model.predict(frame, imgsz=320, conf=0.4, verbose=False)
+        annotated = results[0].plot()
+        yield annotated
+    cap.release()
+
+# 🚀 Gradio App
 with gr.Blocks() as demo:
-    gr.Markdown("# 🚀 Optimized YOLOv8 Detection\nFast & Accurate on Images and Videos")
+    gr.Markdown("# 🔍 YOLOv8 Object Detection")
+    gr.Markdown("Upload an image or video, or use your webcam for real-time detection.")
 
-    with gr.Tab("Image"):
+    with gr.Tab("🖼️ Image Detection"):
         img_input = gr.Image(type="pil")
-        img_output = gr.Image(label="Detected")
-        img_btn = gr.Button("Run Detection")
+        img_output = gr.Image()
+        img_btn = gr.Button("Detect Objects")
         img_btn.click(predict_image, inputs=img_input, outputs=img_output)
 
-    with gr.Tab("Video"):
+    with gr.Tab("🎞️ Video Detection"):
         vid_input = gr.Video()
         vid_output = gr.Video()
-        vid_btn = gr.Button("Run Detection on Video")
+        vid_btn = gr.Button("Detect in Video")
         vid_btn.click(predict_video, inputs=vid_input, outputs=vid_output)
 
-demo.launch()
-
-
+    with gr.Tab("📷 Live Webcam"):
+        gr.Markdown("Note: Webcam streaming may not work on Hugging Face Spaces.")
+        live_img = gr.Image(streaming=True)
+        live_img.stream(live_feed)
 
+demo.launch()