A python template
Project description
video2numpy
Optimized library for large-scale extraction of frames and audio from video.
Install
pip install video2numpy
Or build from source:
python setup.py install
Usage
NAME
video2numpy - Read frames from videos and save as numpy arrays
SYNOPSIS
video2numpy SRC <flags>
DESCRIPTION
Input:
src:
str: path to mp4 file
str: youtube link
str: path to txt file with multiple mp4's or youtube links
list: list with multiple mp4's or youtube links
dest:
str: directory where to save frames to
None: dest = src + .npy
take_every_nth:
int: only take every nth frame
resize_size:
int: new pixel height and width of resized frame
workers:
int: number of workers used to read videos
memory_size:
int: number of GB of shared memory used for reading, use larger shared memory for more videos
POSITIONAL ARGUMENTS
SRC
FLAGS
--dest=DEST
Default: ''
--take_every_nth=TAKE_EVERY_NTH
Default: 1
--resize_size=RESIZE_SIZE
Default: 224
--workers=WORKERS
Default: 1
--memory_size=MEMORY_SIZE
Default: 4
NOTES
You can also use flags syntax for POSITIONAL ARGUMENTS
API
This module exposes a single function video2numpy
which takes the same arguments as the command line tool:
import glob
from video2numpy import video2numpy
VIDS = glob.glob("some/path/my_videos/*.mp4")
FRAME_DIR = "some/path/my_frames"
take_every_5 = 5
video2numpy(VIDS, FRAME_DIR, take_every_5)
You can also directly use the reader and iterate over videos yourself:
import glob
from video2numpy.frame_reader import FrameReader
VIDS = glob.glob("some/path/my_videos/*.mp4")
take_every_5 = 5
resize_size = 300
reader = FrameReader(VIDS, take_every_5, resize_size)
reader.start_reading()
for vid_frames, npy_name in reader:
# do something with vid_frames of shape (n_frames, 300, 300, 3)
...
For development
Either locally, or in gitpod (do export PIP_USER=false
there)
Setup a virtualenv:
python3 -m venv .env
source .env/bin/activate
pip install -e .
to run tests:
pip install -r requirements-test.txt
then
make lint
make test
You can use make black
to reformat the code
python -m pytest -x -s -v tests -k "dummy"
to run a specific test
Project details
Release history Release notifications | RSS feed
Download files
Download the file for your platform. If you're not sure which to choose, learn more about installing packages.
Source Distribution
video2numpy-2.0.0.tar.gz
(9.5 kB
view hashes)
Built Distribution
Close
Hashes for video2numpy-2.0.0-py3-none-any.whl
Algorithm | Hash digest | |
---|---|---|
SHA256 | f1ac47b0e4da081b5edf4088fde1e6484d4e9355405e7802bdee40fd035fefc1 |
|
MD5 | 588d6357b14c89cb5bceb85df1f8026e |
|
BLAKE2b-256 | a19fe40b66547231344e2dee5fbe3c8b39f763590c5d071f3d4a913e15645454 |