Skip to main content

Mechanical Compound Eye

Project description

Mechanical Compound Eye

Is a python based Nvidia DeepStream app to watch multiple video streams at once.

Requirements

Installation

pip3 install (git url)

or

pip3 install mce

or if you don't have pip / don't want to install it

(git clone this url)
cd mce
python3 ./setup.py install

Example Usage:

To run detections on multiple youtube videos/playlists (requires youtube-dl which can be installed with pip3 install youtube-dl):

mce $(youtube-dl -f best -g https://www.youtube.com/watch?v=pJ5cg83D5AE) $(youtube-dl -f best -g https://www.youtube.com/watch?v=peC1JD9gEKc) $(youtube-dl -f best -g https://www.youtube.com/watch?v=0LYE669fbpU)

Any uri supported by uridecodebin should work.

Faq

  • Did you come up with the name? No.
  • How can I customize this? The primary inference config is in ~/.mce/pie.conf

Known Issues

  • tests need to be written
  • the decoder spews warning messages... not sure why
  • videos don't always play at the correct rate. the --live option may help with some sources
  • nvinfer doesn't respect batch-engine-file set at runtime for loading a file, leading to long start times.

Project details


Download files

Download the file for your platform. If you're not sure which to choose, learn more about installing packages.

Source Distribution

mce-0.1.3.tar.gz (25.0 kB view hashes)

Uploaded Source

Supported by

AWS AWS Cloud computing and Security Sponsor Datadog Datadog Monitoring Fastly Fastly CDN Google Google Download Analytics Microsoft Microsoft PSF Sponsor Pingdom Pingdom Monitoring Sentry Sentry Error logging StatusPage StatusPage Status page