Update app.py

app.py

@@ -1,96 +1,53 @@
-import time
-import re
+import cv2
 import gradio as gr
-import torch
-from transformers import pipeline
-from requests_html import HTMLSession
-import speech_recognition as sr
+from ultralytics import YOLO
 
-# Load NLP Model with GPU support
-classifier = pipeline("zero-shot-classification", model="facebook/bart-large-mnli", device=0 if torch.cuda.is_available() else -1)
+# Load YOLOv8 Model
+model = YOLO("yolov8n.pt")  # Use 'yolov8s.pt' for better accuracy
 
-# Define possible intents
-possible_intents = [
-    "open google",
-    "search something",
-    "go to amazon",
-    "order something"
-]
-
-# Function to classify user command
-def get_intent(command):
-    result = classifier(command, possible_intents)
-    intent = result["labels"][0]
-    return intent.lower()
-
-# Function to fetch search results
-def fetch_search_results(url):
-    session = HTMLSession()
-    response = session.get(url)
-    title = response.html.find("title", first=True).text if response.html.find("title", first=True) else "No title found"
-    return title
-
-# Function to execute commands
-def execute_command(command):
-    intent = get_intent(command)
-    output_log = f"Command: {command}\nDetected Intent: {intent}\n"
-
-    if "open google" in intent:
-        output_log += "✅ Google homepage ready.\n"
-
-    elif "search something" in intent:
-        query = re.search(r'search for (.+)', command)
-        if query:
-            search_term = query.group(1)
-            url = f"https://www.google.com/search?q={search_term}"
-            title = fetch_search_results(url)
-            output_log += f"🔍 Searched for: {search_term}. Found: {title}\n"
-
-    elif "go to amazon" in intent:
-        output_log += "🛒 Amazon homepage ready.\n"
-
-    elif "order something" in intent:
-        product = re.search(r'order (.+)', command)
-        if product:
-            search_term = product.group(1)
-            url = f"https://www.amazon.com/s?k={search_term}"
-            title = fetch_search_results(url)
-            output_log += f"🛍️ Ordered: {search_term}. Found: {title}\n"
-    else:
-        output_log += "⚠️ Command not recognized.\n"
-
-    return output_log
-
-# Voice command recognition
-def listen_for_command():
-    recognizer = sr.Recognizer()
-    with sr.Microphone() as source:
-        print("Listening for commands...")
-        audio = recognizer.listen(source)
-        try:
-            command = recognizer.recognize_google(audio)
-            print(f"Command recognized: {command}")
-            return execute_command(command)
-        except sr.UnknownValueError:
-            return "Could not understand the command."
-        except sr.RequestError:
-            return "Error with the speech recognition service."
-
-# Gradio Interface
-with gr.Blocks() as ui:
-    gr.Markdown("# 🔥 Web Automation with NLP + Gradio (GPU Accelerated)")
-
-    with gr.Row():
-        text_input = gr.Textbox(label="Enter Command", placeholder="Example: Search for iPhone on Google")
-        submit_btn = gr.Button("Execute")
-
-    with gr.Row():
-        mic_btn = gr.Button("🎤 Voice Command")
+def detect_cars(video_url):
+    cap = cv2.VideoCapture(video_url)
     
-    output_box = gr.Textbox(label="Execution Log", lines=5)
-
-    submit_btn.click(execute_command, inputs=text_input, outputs=output_box)
-    mic_btn.click(listen_for_command, outputs=output_box)
-
-# Launch Gradio App
-ui.launch()
+    while cap.isOpened():
+        ret, frame = cap.read()
+        if not ret:
+            break
+
+        # Perform YOLO detection
+        results = model(frame)
+
+        # Initialize car count
+        car_count = 0
+
+        # Loop through detections and count cars
+        for result in results:
+            for box in result.boxes:
+                class_id = int(box.cls[0])
+                if class_id in [2, 3, 5, 7]:  # Car-related classes
+                    car_count += 1
+
+        # Draw results on frame
+        frame = results[0].plot()
+
+        # Display car count on screen
+        cv2.putText(frame, f'Cars: {car_count}', (20, 50),
+                    cv2.FONT_HERSHEY_SIMPLEX, 1, (0, 0, 255), 2, cv2.LINE_AA)
+
+        # Convert frame to RGB for display in Hugging Face
+        frame = cv2.cvtColor(frame, cv2.COLOR_BGR2RGB)
+        
+        yield frame
+
+    cap.release()
+
+# Hugging Face Gradio UI
+iface = gr.Interface(
+    fn=detect_cars,
+    inputs="text",
+    outputs="image",
+    live=True,
+    title="YOLOv8 Live Car Detection",
+    description="Enter the IP Webcam URL to detect and count cars in real-time."
+)
+
+iface.launch()