Event-Based Features Selection and Tracking from Intertwined Estimation of Velocity and Generative Contours
This addresses the problem of robust feature tracking in dynamic environments for applications like robotics or autonomous systems, representing a novel method for a known bottleneck in event-based vision.
The paper tackles the problem of detecting and tracking features from event-based camera outputs without relying on predefined shapes like corners, by developing a method that intertwines velocity vector estimation with Bayesian contour description. The result is a system that handles large variations in luminosity, speed, and feature size, achieving very fast convergence rates through speed-based feedback.
This paper presents a new event-based method for detecting and tracking features from the output of an event-based camera. Unlike many tracking algorithms from the computer vision community, this process does not aim for particular predefined shapes such as corners. It relies on a dual intertwined iterative continuous -- pure event-based -- estimation of the velocity vector and a bayesian description of the generative feature contours. By projecting along estimated speeds updated for each incoming event it is possible to identify and determine the spatial location and generative contour of the tracked feature while iteratively updating the estimation of the velocity vector. Results on several environments are shown taking into account large variations in terms of luminosity, speed, nature and size of the tracked features. The usage of speed instead of positions allows for a much faster feedback allowing for very fast convergence rates.