aws-samples / aws-samples/amazon-kinesis-video-streams-consumer-library-for-python
Running the YOLO model in the on_fragment_arrived callback is very slow.
Open
- Dominant language
- Python
- Stars
- 35
- Forks
- 28
- PR merge metrics
- No merged PRs in 30d
Description
I am facing an issue where running the YOLO model detection outside of the on_fragment_arrived callback takes about 100 ms, but running YOLO detection inside the on_fragment_arrived callback takes 2 seconds. Could there be any resource limitations or configurations that are preventing it from utilizing all of the host computer's resources?
Contributor guide
Assessment
This issue has not been assessed yet.