edge-video-services/ekya

Source code and datasets for Ekya, a system for continuous learning on the edge.

37
/ 100
Emerging

Ekya helps organizations continuously improve the accuracy of their video analytics models, especially on devices with limited computing power. It takes live video streams and pre-trained object detection models, then intelligently fine-tunes these models in the background to deliver better detection accuracy over time. This is ideal for operations managers or engineers deploying AI-powered cameras for security, traffic monitoring, or facility management.

113 stars. No commits in the last 6 months.

Use this if you need your video surveillance or analysis systems to adapt and maintain high accuracy on the fly, even when running on edge devices like smart cameras or embedded systems.

Not ideal if your models are static, your data does not drift over time, or you have ample compute resources for traditional, periodic model retraining.

video-analytics edge-computing object-detection surveillance traffic-monitoring
No License Stale 6m No Package No Dependents
Maintenance 0 / 25
Adoption 9 / 25
Maturity 8 / 25
Community 20 / 25

How are scores calculated?

Stars

113

Forks

23

Language

Jupyter Notebook

License

Last pushed

Mar 10, 2022

Commits (30d)

0

Get this data via API

curl "https://pt-edge.onrender.com/api/v1/quality/ml-frameworks/edge-video-services/ekya"

Open to everyone — 100 requests/day, no key needed. Get a free key for 1,000/day.