Event-based Visual Tracking in Dynamic Environments
This work addresses robust object tracking for applications in robotics or autonomous systems under challenging conditions, but it is incremental as it combines existing sensors with off-the-shelf deep learning.
The paper tackles the problem of visual object tracking in dynamic environments with motion blur by using event cameras to reconstruct intensity frames, improving tracking performance where conventional cameras fail.
Visual object tracking under challenging conditions of motion and light can be hindered by the capabilities of conventional cameras, prone to producing images with motion blur. Event cameras are novel sensors suited to robustly perform vision tasks under these conditions. However, due to the nature of their output, applying them to object detection and tracking is non-trivial. In this work, we propose a framework to take advantage of both event cameras and off-the-shelf deep learning for object tracking. We show that reconstructing event data into intensity frames improves the tracking performance in conditions under which conventional cameras fail to provide acceptable results.