Learning course · Updated June 2026
How a modern surveillance system actually works — IP cameras, ONVIF and RTSP, VMS and NVR, edge and cloud video analytics, storage and retention, and privacy compliance. A practical course from Fora Soft engineers, from the camera sensor to the operator's screen.
Every chapter starts with a question and ends with a build decision. Standards cited by name and profile. Privacy tied to named law. No vendor brochures.
Outcomes
Eight blocks that take you from a single IP camera to a federated, compliant, multi-site VMS. By the end, you can specify, build, and operate a video surveillance system that holds up under load — and stands up to an audit.
Pick a path
The same 57 articles, ordered for what you actually need to do this quarter.
From "what is a VMS" to your first end-to-end system. The vocabulary, the camera-to-client anatomy, deployment models, and the cost model.
Make a multi-vendor system work and survive contact with reality. ONVIF, RTSP, the analytics catalogue, the retention math, and storage at scale.
Where analytics run, whether you can legally ship them, which platform to buy, and how to assemble it all per vertical. The engineering and governance behind a production system.
Syllabus
Every chapter is self-contained. Read in order, or jump straight to the block you need — from the camera-to-client anatomy to vertical reference architectures.
01
02
03
04
05
06
07
08
Talk to the engineers who build it. Fora Soft helps teams integrate ONVIF camera fleets, place video analytics on the right tier, size storage, and ship VMS products that hold up — and stay compliant — in the field.
Featured
Hand-picked deep dives across standards, analytics, storage, and compliance — the highest-impact reads first, before you commit to a learning path.
Reference
120+ terms with crisp definitions, aliases, and links to deep dives. From VMS, NVR, and ONVIF to LPR/ANPR, federation, and BIPA — the full A–Z is one click away.
VMS
Video Management System. The software platform that ingests, records, and manages many cameras and recorders across servers — beyond what a single NVR or DVR appliance can do.
NVR
Network Video Recorder. Records the digital streams of IP cameras over a network. The recorder tier beneath a VMS; software NVRs blur the line.
ONVIF
Open Network Video Interface Forum. The standard that lets cameras and VMS platforms from different vendors interoperate — discovery, streaming, PTZ, and metadata.
RTSP
Real-Time Streaming Protocol. The session protocol an IP camera speaks to deliver its RTP video stream into a VMS or NVR.
Video analytics
Software that turns raw video into searchable events — object detection, tracking, license-plate and face recognition, and behavioral and anomaly detection.
Edge AI
On-camera or edge-server analytics that run close to the camera, cutting bandwidth and latency and keeping raw footage local for privacy.
Written and maintained by
FAQ
A DVR (digital video recorder) records analog cameras over coax and is the legacy tier. An NVR (network video recorder) records the digital streams of IP cameras over a network. A VMS (video management system) is the software platform above both — it manages many cameras, recorders, and servers, adds analytics, access control, and multi-site federation, and scales far beyond a single appliance. Most modern builds are VMS-led with IP cameras.
ONVIF (Open Network Video Interface Forum) is the standard that lets IP cameras, NVRs, and VMS platforms from different vendors work together. It standardizes the interfaces — device discovery via WS-Discovery, video and audio streaming, PTZ control, and analytics metadata — through Profiles S, G, T, and M. It does not standardize analytics quality or every vendor feature; for those you fall back to a camera SDK. ONVIF is a baseline, not a ceiling.
It depends on latency, bandwidth, and privacy. Edge analytics — on the camera or an edge server — react in milliseconds, cut upload bandwidth, and keep raw footage local, but the compute is limited. Cloud analytics offer elastic GPU power and easier model updates, at the cost of bandwidth, recurring spend, and data exposure. Most production systems are hybrid: lightweight detection at the edge filters what the cloud analyzes in depth.
There is no single figure — cost scales with camera count, resolution and bitrate, retention period, on-prem versus cloud, and how much analytics compute you run. The big line items are cameras and mounting, storage (the part most projects underestimate), VMS and analytics licenses, servers or cloud, integration, and the ongoing run-rate. Model it per camera per month, not as a one-time number. Chapter 1 ships a cost-model worksheet.
Storage follows simple arithmetic: per-camera bitrate (Mbps) ÷ 8 × 3,600 × recording hours per day × retention days × number of cameras gives gigabytes. H.265 roughly halves the bitrate of H.264 for the same quality, and motion-only recording cuts it further. Retention is usually set by policy or law — Chapter 6 covers the legal caps. Chapter 5 ships a storage and retention worksheet.
Under GDPR (Regulation (EU) 2016/679), CCTV needs a lawful basis, clear signage and notice, data minimization, defined retention limits, and often a Data Protection Impact Assessment for large-scale monitoring. Face recognition adds biometric rules — in Illinois, BIPA requires informed written consent before capturing a faceprint, with a private right of action. Rules vary by region, so design for masking, consent, and lawful retention. This is engineering guidance, not legal advice.
Fora Soft has built real-time video, audio, and AI products since 2005 — WebRTC, LiveKit, generative pipelines, and AI agents at scale. Tell us what you’re building and we’ll send a real engineer your way.
Specialist software house for video, real-time and AI products. Founded 2005. 50 in-house engineers.
