Detection vs classification, the precision/recall/mAP accuracy ranges (never 100%), the motion-vs-AI false-alarm math, where detection runs (edge NPU) and how it surfaces into the VMS via ONVIF Profile M, and the privacy line (a class is not an identity) — on one page.
Download free PDF
The handbook · 7 volumes
Fora Soft Video Engineering Handbook
This object detection in surveillance reference covers detection versus classification, realistic precision and recall, false-alarm math, edge NPUs and ONVIF Profile M. We wrote seven volumes covering the video stack; start with Volume 6 on computer vision, detection, tracking and segmentation, then Volume 7 on real-time pipelines and biometrics rules, with Volume 1 for encoding. Written by Nikolay Sapunov, CEO at Fora Soft.
Cutting false alarms with AI object detection on your cameras?
Share sample footage and your current alarm rates, and we will estimate the false-alarm reduction and recommend where detection should run. 250+ video products shipped since 2005.
You get a realistic estimate, technical recommendations and next steps. Free, no obligations.
