update file

.gitattributes

@@ -33,3 +33,6 @@ saved_model/**/* filter=lfs diff=lfs merge=lfs -text
 *.zip filter=lfs diff=lfs merge=lfs -text
 *.zst filter=lfs diff=lfs merge=lfs -text
 *tfevents* filter=lfs diff=lfs merge=lfs -text
+car.mp4 filter=lfs diff=lfs merge=lfs -text
+openh264-1.8.0-win64.dll filter=lfs diff=lfs merge=lfs -text
+output_video.mp4 filter=lfs diff=lfs merge=lfs -text

car.mp4

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:c7f7287faa231cf287c433ef8b2f38949e214cac9c473aea286d75f2bdea2330
+size 4708643

main.py

@@ -0,0 +1,53 @@
+import cv2
+import gradio as gr
+import validators
+from cap_from_youtube import cap_from_youtube
+
+import video_detection
+from video_detection import video_detection
+
+
+def preprocess_input(input):
+    # if not input:
+    #     cap = cv2.VideoCapture(0)
+    #     yield from video_detection(cap)
+    if validators.url(input):
+        if 'youtu' in input:
+            cap = cap_from_youtube(input, resolution='720p')
+            yield from video_detection(cap)
+        else:
+            print("Invalid URL")
+    else:
+        cap = cv2.VideoCapture(input)
+        yield from video_detection(cap)
+
+# gradio interface
+input_video = gr.Video(label="Input Video")
+input_url = gr.Textbox(label="Input URL", placeholder="Enter URL")
+output_frames_1 = gr.Image(label="Output Frames")
+output_video_file_1 = gr.Video(label="Output video")
+output_frames_2 = gr.Image(label="Output Frames")
+output_video_file_2 = gr.Video(label="Output video")
+# sample_video=r'sample/car.mp4'
+
+file_tab = gr.Interface(
+    fn=preprocess_input,
+    inputs=[input_video],
+    outputs=[output_frames_1, output_video_file_1],
+    title=f"Завантажте файл для розпізнавання",
+    allow_flagging="never",
+    examples=[["car.mp4"]],
+)
+
+url_tab = gr.Interface(
+    fn=preprocess_input,
+    inputs=[input_url],
+    outputs=[output_frames_2, output_video_file_2],
+    title=f"Введіть URL Youtube відео для розпізнавання",
+    allow_flagging="never",
+    examples=[["car.mp4"]],
+)
+
+app = gr.TabbedInterface([file_tab, url_tab], ["Завантажити файл", "Ввести URL"])
+
+app.launch()

openh264-1.8.0-win64.dll

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:e3a5cf05b673a17ebfe95ac6b479607cf4df2289fe4a6d5f3d3ff09aa8f56192
+size 825160

output_video.mp4

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:4d28471513da5bc1cba0b90a35965c665a52804d456185201c2c36e4ef2022dc
+size 29715980

video_detection.py

@@ -0,0 +1,58 @@
+from ultralytics import YOLO
+import cv2
+
+model = YOLO("models/yolov8m.pt")
+
+def draw_boxes(frame, results):
+    for r in results:
+        boxes = r.boxes
+
+        for box in boxes:
+            x1, y1, x2, y2 = box.xyxy[0]
+            x1, y1, x2, y2 = int(x1), int(y1), int(x2), int(y2)
+
+            cv2.rectangle(frame, (x1, y1), (x2, y2), (255, 0, 255), 3)
+
+            cls = r.names[box.cls[0].item()]
+
+            # object details
+            org = [x1, y1]
+            font = cv2.FONT_HERSHEY_SIMPLEX
+            fontScale = 1
+            color = (255, 0, 0)
+            thickness = 2
+
+            cv2.putText(frame, cls, org, font, fontScale, color, thickness)
+
+    return frame
+
+
+def video_detection(cap):
+    frame_width = int(cap.get(cv2.CAP_PROP_FRAME_WIDTH))
+    frame_height = int(cap.get(cv2.CAP_PROP_FRAME_HEIGHT))
+    fps = int(cap.get(cv2.CAP_PROP_FPS))
+
+    out = cv2.VideoWriter('output_video.mp4', cv2.VideoWriter_fourcc(*'h264'), fps, (frame_width, frame_height))
+
+    count = 0
+    while cap.isOpened():
+        success, frame = cap.read()
+
+        if not success:
+            break
+
+        results = model(frame, stream=True, device='cuda', verbose=False)
+
+        frame = draw_boxes(frame, results)
+
+        out.write(frame)
+        if not count % 10:
+            yield frame, None
+        # print(count)
+        count += 1
+
+    cap.release()
+    out.release()
+    cv2.destroyAllWindows()
+
+    yield None, 'output_video.mp4'