Update inference.py

inference.py

@@ -2,13 +2,11 @@ import time
 import cv2
 import numpy as np
 import onnxruntime
-
 from utils import draw_detections
 
 
 class YOLOv10:
     def __init__(self, path):
-
         # Initialize model
         self.initialize_model(path)
 
@@ -53,7 +51,7 @@ class YOLOv10:
         )
 
         print(f"Inference time: {(time.perf_counter() - start)*1000:.2f} ms")
-        boxes, scores, class_ids, = self.process_output(outputs, conf_threshold)
+        boxes, scores, class_ids = self.process_output(outputs, conf_threshold)
         return self.draw_detections(image, boxes, scores, class_ids)
 
     def process_output(self, output, conf_threshold=0.3):
@@ -82,9 +80,6 @@ class YOLOv10:
         # Scale boxes to original image dimensions
         boxes = self.rescale_boxes(boxes)
 
-        # Convert boxes to xyxy format
-        #boxes = xywh2xyxy(boxes)
-
         return boxes
 
     def rescale_boxes(self, boxes):
@@ -117,30 +112,35 @@ class YOLOv10:
 
 
 if __name__ == "__main__":
-    import requests
-    import tempfile
     from huggingface_hub import hf_hub_download
 
     model_file = hf_hub_download(
         repo_id="onnx-community/yolov10s", filename="onnx/model.onnx"
     )
 
-    yolov8_detector = YOLOv10(model_file)
+    yolov10_detector = YOLOv10(model_file)
 
-    with tempfile.NamedTemporaryFile(suffix=".jpg", delete=False) as f:
-        f.write(
-            requests.get(
-                "https://live.staticflickr.com/13/19041780_d6fd803de0_3k.jpg"
-            ).content
-        )
-        f.seek(0)
-        img = cv2.imread(f.name)
+    cap = cv2.VideoCapture(1)
+
+    if not cap.isOpened():
+        print("Error: Could not open video stream.")
+        exit()
+
+    while True:
+        ret, frame = cap.read()
+        if not ret:
+            print("Failed to grab frame. Exiting...")
+            break
+
+        # Perform object detection on the frame
+        detected_frame = yolov10_detector.detect_objects(frame)
 
-    # # Detect Objects
-    combined_image = yolov8_detector.detect_objects(img)
+        # Display the frame with detections
+        cv2.imshow("Mobile Camera Feed with YOLOv10", detected_frame)
 
+        # Press 'q' to quit
+        if cv2.waitKey(1) & 0xFF == ord("q"):
+            break
 
-    # Draw detections
-    cv2.namedWindow("Output", cv2.WINDOW_NORMAL)
-    cv2.imshow("Output", combined_image)
-    cv2.waitKey(0)
+    cap.release()
+    cv2.destroyAllWindows()