Careers
Our software runs in airports, city traffic systems, defence control rooms and factories. Read about the work and how to apply.
Videolytical Systems Pvt. Ltd. is a software company in Noida, India. We build AI video analytics and video management software.
A site with many cameras produces more video than people can watch. Our software analyses that video and tells security and operations teams about events as they happen.
Our video analytics run AI and deep-learning models on live and recorded video. They extract metadata and send real-time alerts to command and control centres. We also build video management software for live view, recording and playback, and video cloud software for estates with more than one site.
The software comes as five products that work alone or together. The same stack runs at the edge, on-premise or in the cloud. It takes video from existing IP cameras and network video recorders over ONVIF and RTSP, so it fits the infrastructure a site already owns.
The software reaches sites through partners: original equipment manufacturers (OEMs) of cameras and hardware, security solution providers and system integrators.
Three convictions shape how we build the software and how we run a project.
We start from the problems of the site, not from a product list. A pilot on a small set of cameras shows what the analytics do on your own scenes. The wider rollout follows once the pilot has shown its value.
Real sites bring glare, dust, monsoon rain, low light, dense crowds and cameras of mixed quality. Our models are trained and hardened for these conditions.
The software is designed, engineered and supported from one office in Noida. Engineering and support work from the same place.
A project runs in five steps, from the first requirement to a running deployment.
We work with your team to map the security and analytics needs of the site, such as what to detect, on which cameras and who receives the alerts. This step typically ends with an agreed scope for the project.
Our architects plan where the analytics will run: at the edge, on-premise, in the cloud or a mix of the three. The plan is built around the cameras, network and systems you already have. The result is a sizing and deployment plan for the site.
Deep-learning models are selected for the events you need. Where a site has objects of its own, a model can be trained for them. Each model gets its alert rules: zones, schedules and confidence thresholds for each camera.
The system is deployed with our partners, then configured, calibrated and connected to your command centre. Deployment typically starts with a pilot on a small set of cameras and scales out from there.
The system is tested for accuracy, performance and reliability. It is then handed over to your team with training. Optimisation continues in the field after handover.
We work with three kinds of partner. A customer may meet our software inside a device, inside a wider security solution, or in a project run by an integrator.
Address
277, C Block, Sector 63, Noida, Uttar Pradesh, India
Write to sales for demos, sample clips and pilots. Write to info for everything else.
Our software runs in airports, city traffic systems, defence control rooms and factories. Read about the work and how to apply.
A company profile, brand assets and the contact for media enquiries.
Practical guides from our engineering team for buyers and operators of video systems.
How to get help with a deployment, and answers to common questions.
Arrange a live demo, send a sample clip, or ask for an architecture and sizing discussion for your site.