Enhancing Event-Based Optical Camera Communication Via Dynamic Timing Correction

Matthew Howard, Keigo Hirakawa; Proceedings of the IEEE/CVF International Conference on Computer Vision (ICCV) Workshops, 2025, pp. 4656-4664

Event-based optical sensors have broken the conventional fixed sample rate paradigm of high-speed imaging that normally imposes a harsh limit on sample rate and number of pixels, effectively offering megapixel sensors capable of measuring at 10s of kHz. This behavior is well suited for optical camera communications as has been proposed recently. We use principles of the hardware operation to design an EB-OCC method that improves upon prior works by 25 times and facilitates more computationally efficient methods of beacon extraction and tracking. Results presented utilize well calibrated signals in a laboratory environment to aid prediction of performance in the real world. The proposed method achieves up to 75 kbps channel capacity and can compensate for scene motion up to at least the 60 deg/second we reliably generate in the lab.