Glossary

Check (in Svid)

A check in Svid is a paid watcher over the core pipeline: alerting checks judge recent frames and raise alerts, report checks read whole videos and write documents.

Dwell time

Dwell time is how long a person stays in a defined area, from entering to leaving. Read as engagement in retail, and as a warning sign in safety.

Footfall

Footfall is the number of people entering or passing through a space over a given period, and it's the basic vitality metric of any physical shop.

Motion gating

Motion gating skips video frames that barely differ from the last kept frame before any AI runs, so mostly-static CCTV footage costs a fraction of busy footage.

Natural language video search

Natural language video search finds moments in footage from a typed description, like “person in a red jacket near the till”, with no scrubbing or timestamps.

Re-identification (re-ID)

Re-identification (re-ID) is matching the same person across cameras or over time without knowing who they are. Svid deliberately does no re-identification.

RTSP

RTSP (Real Time Streaming Protocol) is the standard way IP cameras expose a live video stream so other software, like AI video analysis, can read it.

Searchable video

Searchable video is footage an AI has described frame by frame, so you can query it in plain English. It's also what Svid stands for: Searchable Video.

Shrinkage (retail)

Shrinkage is the stock a retailer loses between delivery and sale, through theft, error or damage. Most of it happens on camera and goes unwatched.

Video analytics

Video analytics is the automated analysis of video by software, traditionally to detect motion, count objects or flag events like a line being crossed.

Video intelligence

Video intelligence is the use of AI to extract meaning from video, the people, actions and events in footage, rather than just recording and storing it.

Video retention

Video retention is how long footage is kept before deletion. Why operators pick 14, 30 or 90 days, what UK ICO guidance expects, and where AI-derived data fits.

Vision-language model (VLM)

A vision-language model (VLM) is an AI model that looks at an image and describes what it sees in words, the engine that makes video footage searchable.

VMS (video management system)

A VMS (video management system) records camera feeds, stores the footage and plays it back. It indexes time and motion, but not what actually happened.