• • The event-driven sensor achieves a gesture recognition accuracy exceeding 92.5%, enabling reliable UAV control in real-time HMI scenarios.
• • In a simulated 10-second scene at 1920×1080 resolution, the event-based sensor generates an average file size of 7.5 MB, compared to 14.8 MB for a frame-based sensor, representing a 49.3% reduction in data redundancy.
• • The device architecture (Sb-doped TiO2/QDs/PEDOT:PSS) is solution-processed and two-terminal, simplifying fabrication and potentially lowering production costs compared to complex DVS circuits.
• • The hybrid HMI system combines event-driven sensing for low-latency gesture control with frame-based sensing for high-precision gaze control, demonstrating a scalable strategy for multi-stage tasks.