The client needed a scalable solution to deliver real-time video intelligence and new features without forcing their customers to upgrade existing camera hardware.
Delivered a scalable, cloud-agnostic video intelligence platform with legacy camera support and an optional turnkey hardware solution.
They could now deliver real-time analytics and new features to existing customer cameras - without costly hardware upgrades - while also offering a turnkey onprem/cloud deployment option.
- Enable existing ONVIF-compatible cameras to become smarter through cloud-delivered visual intelligence.
- Offer a fully turnkey, white-labeled hardware-software solution for customers wanting an out-of-the-box experience.
NitroCam aimed to become the “brain” for any camera, making software-defined surveillance accessible at scale.
NitroCam needed a scalable, cloud-based visual intelligence solution that could transform existing surveillance systems while also providing a hardware-based reference design for new deployments. The platform had to work on-premise and in the cloud, deliver real-time video analytics, and maintain strong data privacy protections while still enabling remote support and troubleshooting.
- Built a containerized, cloud-agnostic video streaming architecture.
- Delivered adaptive storage layers to support varying service models.
- Enabled real-time analytics and enterprise-level security for legacy ONVIF-compatible cameras.
- Supported both on-premise and cloud deployments.
For customers preferring a bundled solution or in cases where they didn't already have cameras:
- Designed a camera based on the Ambarella S5 chipset, providing a ready-to-deploy hardware option alongside the software platform.
- Developed custom firmware on the S5 chipset to ensure seamless plug-and-play functionality.
- Overcame ODM support limitations and third-party library porting challenges.
- Optimized for compute-constrained environments by replacing heavy libraries with lighter alternatives.
- Optimized the conventional complex camera configuration to zero config boot up.
- Created a responsive web and mobile interface
- Built native iOS and Android apps supporting QR code-based setup and cloud integration
- Ran focus groups with early adopters to continuously refine UI/UX
- Engineered a stream-separated logging system to minimize bandwidth usage
- Enabled scalable backend processing for device diagnostics
- Maintained strict user-role isolation, ensuring customer data privacy while allowing remote troubleshooting
The result was a modular, scalable, high-performance multi-core framework capable of real-time video analytics and enterprise-scale deployment, whether on-premise or in the cloud.
- Real-Time Alerts and Events:
Smart visual intelligence apps enabled automatic detection and logging of key events, including fall detection, theft detection, and other customizable security triggers tailored to customer needs. - Forensic and Event Search:
Advanced filters and proprietary fingerprinting algorithms allowed searching petabytes of video in seconds. - Business Intelligence Reporting:
Users could generate dashboards, reports, and actionable insights on product placement, customer movement, and access patterns.
- Extended the lifespan and value of customers' existing camera infrastructure by delivering cloud-based intelligence without requiring hardware upgrades.
- Enabled NitroCam to roll out new features and intelligence apps (like fall detection and theft detection) without forcing clients to replace their existing cameras.
- Offered a hardware reference design, giving NitroCam a turnkey hardware-software deployment option.
- Delivered a scalable, cloud-agnostic streaming backend, helping NitroCam avoid third-party streaming costs.
- Enhanced provisioning and diagnostics workflows through mobile apps and backend tools.
- Maintained enterprise-level privacy controls while still allowing efficient remote troubleshooting and support.
