Chess Dynamics has announced the release of Deep Embedded Feature Tracking (DEFT), an advanced real-time video tracking system that delivers precise and resilient tracking performance even in intricate scenarios.
DEFT employs an advanced deep learning methodology pioneered by the company’s Vision4ce brand, establishing an extensive model of the tracked target. This enables the system to achieve precise target localization and consistently reacquire the target after periods of obstruction. DEFT significantly enhances tracking performance, particularly for challenging targets exhibiting varying appearances or rapid acceleration, even amidst foreground and background distractions, where conventional algorithms often encounter difficulties.
This technology amplifies Chess Dynamics’ AI-powered target detection and tracking prowess, seamlessly integrating with Neural Network-based object detection and target classification for various entities, including multi-rotor drones, vessels, and land vehicles.
As the tracking process unfolds, the model undergoes continuous refinement, bolstering its grasp of the target. This refinement leads to consistently precise and durable tracking performance over the long term. The system’s precise classification capabilities enable it to harness automation technology for threat prioritization and alerts. Additionally, it generates robust analytics data, empowering more accurate decision-making processes.
David Tuddenham, Chess Group Managing Director, emphasized the growing threat posed by stealthy drones and adaptable forces to security and privacy. These elusive devices exploit complex environments to elude conventional surveillance methods, underscoring the pressing demand for innovative technologies to counter these threats. (Press Release)
Image Credit: Chess Dynamics