mextractor can extract media metadata to YAML and read them


Keywords
metadata, video, image, bigdata
License
Apache-2.0
Install
pip install mextractor==5.0.0

Documentation

mextractor: media metadata extractor

Videos and images can be large.

Installation

Download and install from PyPi with pip:

pip install mextractor

If you are extracting metadata from videos, install additional dependencies:

pip install mextractor[video-extract]

Usage

Please back up your files before using them with the package, things might break during runtime causing corruption.

Command line interface (CLI)

Copy directory to a new directory while extracting media info and a single frame from videos in subdirectories:

mextractor video-subdirs <path_to_root>

Programmatically

These functions are useful when integrating mextractor to your own package. You can also use it for quick scripts, see the mextractor.workflows submodule for inspiration.

Extract and dump metadata

Video
from mextractor.workflow import extract_and_dump_video

metadata = extract_and_dump_video(dump_dir, path_to_video, include_image, greyscale, lossy_compress_image)
Image
from mextractor.workflow import extract_and_dump_image

metadata = extract_and_dump_image(dump_dir, path_to_image, include_image, greyscale, lossy_compress_image)

Load media

Video
import mextractor

video_metadata = mextractor.load(mextractor_dir)

print(video_metadata.average_fps)
print(video_metadata.frames)
print(video_metadata.resolution)
print(video_metadata.video_length_in_seconds)
Image
import mextractor

image_metadata = mextractor.load(mextractor_dir)

print(image_metadata.resolution)