Utils for computer vision
pip install rutils
from argparse import ArgumentParser
from rutils.common import str2path
opt_parser = ArgumentParser()
opt_parser.add_argument('--txt_path', type=str2path)
opt = opt_parser.parse_args()
print(type(opt.txt_path))
from rutils.common import run_command
command = 'echo "Run command example"'
run_command(command)
video_path = Path('./test.mp4')
video_reader = rutils.video.VideoReader(in_video_path=video_path)
print(video_reader.num_frames)
for _ in range(num_frames):
frame = video_reader.get_next_frame()
Support two method:
- Use OpenCV
video_writer = rutils.video.VideoWriter(
out_path,
video_w,
video_h,
fps=25,
encoding='H264',
video_bitrate='11M',
mode='opencv'
)
video_writer.write_frame(frame)
video_writer.release()
- Use FFMpeg
video_writer = rutils.video.VideoWriter(
out_path,
video_w,
video_h,
fps=25,
encoding='H264',
video_bitrate='11M',
mode='ffmpeg'
)
video_writer.write_frame(frame)
video_writer.release()