Uploaded May 2021 | Updated September 2026, 2 weeks ago
Source: github.com/WongKinYiu/PyTorch_YOLOv4
Download my 4k video test sequence: archive.org/details/0002201705192
Command:
python detect.py --source ~/Videos/0002-20170519-2.mp4 --half --img-size 1024 --output tmp/
Output:
ONNX_EXPORT: False
Namespace(agnostic_nms=False, augment=False, cfg='cfg/yolov4-pacsp.cfg', classes=None, conf_thres=0.3, device='', fourcc='mp4v', half=True, img_size=1024, iou_thres=0.6, names='data/coco.names', output='tmp', save_txt=False, source='/home/karol/Videos/0002-20170519-2.mp4', view_img=False, weights='weights/yolov4-pacsp.pt')
Using CUDA device0 _CudaDeviceProperties(name='GeForce RTX 2070 with Max-Q Design', total_memory=7981MB)
Model Summary: 342 layers, 5.29214e+07 parameters, 5.29214e+07 gradients
video 1/1 (1/45922) /home/karol/Videos/0002-20170519-2.mp4: 576x1024 4 persons, 9 cars, 1 buss, Done. (0.092s)
video 1/1 (2/45922) /home/karol/Videos/0002-20170519-2.mp4: 576x1024 4 persons, 9 cars, 1 buss, Done. (0.021s)
video 1/1 (3/45922) /home/karol/Videos/0002-20170519-2.mp4: 576x1024 5 persons, 9 cars, 1 buss, 1 traffic lights, Done. (0.021s)
video 1/1 (4/45922) /home/karol/Videos/0002-20170519-2.mp4: 576x1024 5 persons, 9 cars, 1 buss, Done. (0.030s)
video 1/1 (5/45922) /home/karol/Videos/0002-20170519-2.mp4: 576x1024 5 persons, 9 cars, 1 buss, Done. (0.021s)
video 1/1 (6/45922) /home/karol/Videos/0002-20170519-2.mp4: 576x1024 4 persons, 9 cars, 1 buss, 1 trucks, Done. (0.021s)
video 1/1 (7/45922) /home/karol/Videos/0002-20170519-2.mp4: 576x1024 6 persons, 9 cars, 1 buss, Done. (0.020s)
video 1/1 (8/45922) /home/karol/Videos/0002-20170519-2.mp4: 576x1024 6 persons, 9 cars, 1 buss, Done. (0.021s)
video 1/1 (9/45922) /home/karol/Videos/0002-20170519-2.mp4: 576x1024 7 persons, 9 cars, 1 buss, Done. (0.021s)
video 1/1 (10/45922) /home/karol/Videos/0002-20170519-2.mp4: 576x1024 6 persons, 8 cars, 1 buss, Done. (0.023s)
video 1/1 (11/45922) /home/karol/Videos/0002-20170519-2.mp4: 576x1024 6 persons, 8 cars, 1 buss, 1 handbags, Done. (0.020s)
video 1/1 (12/45922) /home/karol/Videos/0002-20170519-2.mp4: 576x1024 6 persons, 8 cars, 1 buss, 1 handbags, Done. (0.021s)
video 1/1 (13/45922) /home/karol/Videos/0002-20170519-2.mp4: 576x1024 6 persons, 8 cars, 1 buss, 1 handbags, Done. (0.079s)
video 1/1 (14/45922) /home/karol/Videos/0002-20170519-2.mp4: 576x1024 6 persons, 8 cars, 1 buss, 1 handbags, Done. (0.021s)
video 1/1 (15/45922) /home/karol/Videos/0002-20170519-2.mp4: 576x1024 6 persons, 8 cars, 1 buss, 1 handbags, Done. (0.020s)
video 1/1 (16/45922) /home/karol/Videos/0002-20170519-2.mp4: 576x1024 6 persons, 8 cars, 1 buss, 2 handbags, Done. (0.024s)
Source: github.com/WongKinYiu/PyTorch_YOLOv4
Download my 4k video test sequence: archive.org/details/0002201705192
Command:
python detect.py --source ~/Videos/0002-20170519-2.mp4 --half --img-size 1024 --output tmp/
Output:
ONNX_EXPORT: False
Namespace(agnostic_nms=False, augment=False, cfg='cfg/yolov4-pacsp.cfg', classes=None, conf_thres=0.3, device='', fourcc='mp4v', half=True, img_size=1024, iou_thres=0.6, names='data/coco.names', output='tmp', save_txt=False, source='/home/karol/Videos/0002-20170519-2.mp4', view_img=False, weights='weights/yolov4-pacsp.pt')
Using CUDA device0 _CudaDeviceProperties(name='GeForce RTX 2070 with Max-Q Design', total_memory=7981MB)
Model Summary: 342 layers, 5.29214e+07 parameters, 5.29214e+07 gradients
video 1/1 (1/45922) /home/karol/Videos/0002-20170519-2.mp4: 576x1024 4 persons, 9 cars, 1 buss, Done. (0.092s)
video 1/1 (2/45922) /home/karol/Videos/0002-20170519-2.mp4: 576x1024 4 persons, 9 cars, 1 buss, Done. (0.021s)
video 1/1 (3/45922) /home/karol/Videos/0002-20170519-2.mp4: 576x1024 5 persons, 9 cars, 1 buss, 1 traffic lights, Done. (0.021s)
video 1/1 (4/45922) /home/karol/Videos/0002-20170519-2.mp4: 576x1024 5 persons, 9 cars, 1 buss, Done. (0.030s)
video 1/1 (5/45922) /home/karol/Videos/0002-20170519-2.mp4: 576x1024 5 persons, 9 cars, 1 buss, Done. (0.021s)
video 1/1 (6/45922) /home/karol/Videos/0002-20170519-2.mp4: 576x1024 4 persons, 9 cars, 1 buss, 1 trucks, Done. (0.021s)
video 1/1 (7/45922) /home/karol/Videos/0002-20170519-2.mp4: 576x1024 6 persons, 9 cars, 1 buss, Done. (0.020s)
video 1/1 (8/45922) /home/karol/Videos/0002-20170519-2.mp4: 576x1024 6 persons, 9 cars, 1 buss, Done. (0.021s)
video 1/1 (9/45922) /home/karol/Videos/0002-20170519-2.mp4: 576x1024 7 persons, 9 cars, 1 buss, Done. (0.021s)
video 1/1 (10/45922) /home/karol/Videos/0002-20170519-2.mp4: 576x1024 6 persons, 8 cars, 1 buss, Done. (0.023s)
video 1/1 (11/45922) /home/karol/Videos/0002-20170519-2.mp4: 576x1024 6 persons, 8 cars, 1 buss, 1 handbags, Done. (0.020s)
video 1/1 (12/45922) /home/karol/Videos/0002-20170519-2.mp4: 576x1024 6 persons, 8 cars, 1 buss, 1 handbags, Done. (0.021s)
video 1/1 (13/45922) /home/karol/Videos/0002-20170519-2.mp4: 576x1024 6 persons, 8 cars, 1 buss, 1 handbags, Done. (0.079s)
video 1/1 (14/45922) /home/karol/Videos/0002-20170519-2.mp4: 576x1024 6 persons, 8 cars, 1 buss, 1 handbags, Done. (0.021s)
video 1/1 (15/45922) /home/karol/Videos/0002-20170519-2.mp4: 576x1024 6 persons, 8 cars, 1 buss, 1 handbags, Done. (0.020s)
video 1/1 (16/45922) /home/karol/Videos/0002-20170519-2.mp4: 576x1024 6 persons, 8 cars, 1 buss, 2 handbags, Done. (0.024s)
