Starstruk Logo MarkStarstruk

v0.2.0

blakeblackshear/frigatev0.2.0Jul 31, 2019by blakeblackshear

AI Summary

Introduces hardware acceleration to significantly reduce CPU usage and improves configuration options for video processing.

Key Highlights

  • Hardware accelerated video decoding via FFMPEG sub-process (60-70% CPU reduction)
  • New `take_frame` option to reduce framerates when supported
  • Improved label positioning to avoid overlapping with detected objects
  • Inclusion of object area in labels to help determine min_person_area
  • Greatly reduced Docker image size from ~2GB to 450MB

New Features

  • Hardware accelerated decoding
  • `take_frame` option
  • Label positioning improvements
  • Smaller Docker image size
  • Benchmarking script

Full Release Notes

- Video decoding is now done in an FFMPEG sub process which enables hardware accelerated decoding of video streams. For me, this reduced CPU usage for decoding by 60-70%. (Fixes #21)
- New `take_frame` option to reduce framerates with frigate when the camera doesnt support it (Fixes #40)
- Tweaked the position of the labels to avoid overlapping with detected objects (Fixes #39)
- Added the area of the object to the label to help determine min_person_area values (thanks @aav7fl)
- Greatly reduced Docker image size, from ~2GB to 450MB
- Added support for custom Odroid-XU4 build (unfortunately, I wasn't able to get the Coral performance to be good enough for me with this board)
- Latest Coral drivers from Google
- Added a benchmarking script to test inference times
- Added some comments to better document config options (Fixes #46)