Real-time Tracking Based on Neuromrophic Vision
This work enables faster real-time tracking for applications using neuromorphic vision sensors compared to conventional cameras, but it is incremental as it adapts existing methods to a new sensor type.
The paper tackled real-time object tracking by combining computer vision algorithms with neuromorphic event-based sensors, achieving tracking at 100 time bins per second.
Real-time tracking is an important problem in computer vision in which most methods are based on the conventional cameras. Neuromorphic vision is a concept defined by incorporating neuromorphic vision sensors such as silicon retinas in vision processing system. With the development of the silicon technology, asynchronous event-based silicon retinas that mimic neuro-biological architectures has been developed in recent years. In this work, we combine the vision tracking algorithm of computer vision with the information encoding mechanism of event-based sensors which is inspired from the neural rate coding mechanism. The real-time tracking of single object with the advantage of high speed of 100 time bins per second is successfully realized. Our method demonstrates that the computer vision methods could be used for the neuromorphic vision processing and we can realize fast real-time tracking using neuromorphic vision sensors compare to the conventional camera.