https://github.com/tux-o-matic/hems

https://github.com/tux-o-matic/hems

Last synced: about 5 hours ago
JSON representation

• Host: GitHub
• URL: https://github.com/tux-o-matic/hems
• Owner: tux-o-matic
• License: gpl-3.0
• Created: 2018-04-23T18:44:29.000Z (over 6 years ago)
• Default Branch: master
• Last Pushed: 2022-06-22T13:18:12.000Z (over 2 years ago)
• Last Synced: 2024-08-17T00:28:58.069Z (3 months ago)
• Language: Python
• Size: 70.8 MB
• Stars: 0
• Watchers: 2
• Forks: 0
• Open Issues: 0
• Metadata Files:
  • Readme: README.md
  • License: LICENSE

Awesome Lists containing this project

README

        
# hems

## Centralized object detection for video streams

### Hardware

Focuses on Raspberry Pi, can run on different hardware with the right packages and h264 encoder.

### Launching stream on Pi with cam (source)

```shell

ffmpeg -i /dev/video0 -vcodec h264_omx -pix_fmt yuv420p -f rawvideo -vf scale=1280:720 -copyts -start_at_zero -r 15 -g 30 -b:v 1M -bufsize 1M -an  -payload_type 96 -max_delay 5000 -f rtp 'udp://224.1.1.2:5000'

```

or to test in VLC:

```shell

ffmpeg -i /dev/video0 -vcodec h264_omx -pix_fmt yuv420p -f rawvideo -vf scale=1280:720 -copyts -start_at_zero -r 15 -g 30 -b:v 1M -bufsize 1M -an -flags global_header -payload_type 96 -max_delay 5000 -f rtp 'udp://224.1.1.2:5000'

```

### Credits

pyimagesearch.com for OpenCV examples