Codex AI Vision

The ultimate privacy-friendly video analytics solution

Introduction

Codex AI Vision is a computer vision solution dedicated to real time video analytics, that leverages the latest advances in AI to assist urban CCTV’s operators in their day-to-day tasks. The solution offers multiple capabilities including real-time people and object tracking across large live video streams.

Download brochure
Download technical sheet

Enhance capabilities & efficiency of video security operator

Research suggests that they suffer from fatigue after only 18 minutes, leading to poor surveillance performance and missed abnormal events. Codex AI Vision allows to track or search a person of interest (POI) across a network of cameras, allowing security operators to improve their efficiency in their daily missions for safety & security.

Our features

 

Live investigation

 

Codex AI Vision allows to track or search a person of interest (POI) across a network of cameras, allowing security operators to improve their efficiency in their daily missions for safety & security such as ensuring safety & security for eldery people, keeping civilians safe, searching for a missing person based on a sparse description and tracing in live streams for a person etc…

  Track, trace & search

 

Track and trace POI – Track person of interest over camera network. Follow active tracking from live camera tab, get notifications for each POI apparition, see the route being updated live on the map.
Search POI: search for a person of interest directly on live streams or recent past (last 5, 10, … min).

Search – It can be based on sparse description of attributes (top, bottom, …) or by photo. Start immediate tracking from search results or add search results information to ongoing case.

Person identification

 

Person re-identification (Re-ID) aims at retrieving a person of interest across multiple non-overlapping cameras. Codex AI Vision can find and re-identify an individual throughout his journey, whether they are in front, in profile or from back. Atos computer vision solution will find all the corresponding silhouettes from the video feeds of a selected area. The search mechanism is not based on a simple object classification, but on a digital signature (non-biometric, non-unique and not allowing the identification of a person) allowing to find an individual with the camera network and to re-identify this person in another area.

Our differentiators

Compatibility with existing infrastructure

Scalable architecture to fit growing needs

Compliance with GDPR regulations

Cutting-edge body recognition

An international recognized expertise

10+ patents in deep learning-based
computer vision and tracking

5+ publications in international
conferences

Experts team in deep learning model
design & optimization

Codex AI Vision cited in Gartner’s latest
study “Emerging technologies: Edge AI
Innovator 2020

Contact us for more information on Codex AI Vision