AI & Video Analytics

Video Analytics features are at the core of the NSoft Vision system. All AI features are fully developed and tested in-house to bring added value to surveillance.

16 articles in this collection

Written by

What are the AI services?

There are two main case aspects that this system covers. The first one is a classical surveillance aspect that includ...

Written by NSoft Vision

Updated

What is a face detection?

Face detection is a feature independent of head detection, where inside a subframe of the detected head system finds ...

Written by NSoft Vision

Updated

What is a head detection?

Head detection is a core AI feature and an initial phase executed in a Vision system. Head detection execution is ind...

Written by NSoft Vision

Updated

What is a people count?

People count is a feature showing how many people are currently positioned in front of the camera. People count is ...

Written by NSoft Vision

Updated

What is sighting?

Sighting is a collection of sequential detections in one uninterrupted appearance in front of the camera with AI proc...

Written by NSoft Vision

Updated

How much can sighting last?

The sighting duration is in a proportional relation with the time period in which the system is performing detections...

Written by NSoft Vision

Updated

What are sightings w/o identity?

As the name implies, this is a sighting recorded by the person's appearance but is not ascribed to an identity. Sight...

Written by NSoft Vision

Updated

What is sightings thumbnail?

A column named "Sighting" is located inside an identity profile as part of the directory. Each registered sighting in...

Written by NSoft Vision

Updated

What is the data distance? (Central embeder)

Each identity is saved in a cloud database in the form of a vector, numerically marked with 512 digits. Central embed...

Written by NSoft Vision

Updated

The system did not recognise me. What is the reason?

Face coverage - System makes an estimation about the face from doing face attribute analysis. If one of those importa...

Written by NSoft Vision

Updated

Identities mix with each other in my space. How can I prevent this?

Assuming you have positioned your cameras appropriately, in such cases, if the two identities are distinctly similar ...

Written by NSoft Vision

Updated

Is it possible to merge identities? How is it done?

The merge feature was built with the purpose of reducing duplicated or surplus identities that are actually presentin...

Written by NSoft Vision

Updated

What happens in the backend when two identities are merged?

Each detected identity is stored in a database on a central platform as a vector. Each vector is unique and is a dire...

Written by NSoft Vision

Updated

What does face recognition process look like once a person appears in front of the camera?

Once identity walks in front of the camera, the system goes through certain face recognition steps. These are not act...

Written by NSoft Vision

Updated

What do horizontal and vertical lines around nose and eyes do?

Horizontal and vertical lines in the eyes and nose area serve as a visualization mark signalizing that head orientati...

Written by NSoft Vision

Updated

How is age and gender prediction done?

Behind the age and gender prediction lies the work of the neural network or a "model" that was trained with the purpo...

Written by NSoft Vision

Updated