Iterative Event-based Motion Segmentation by Variational Contrast Maximization
This work addresses motion segmentation for event-based vision systems, useful in applications like object detection and visual servoing, but it is incremental as it builds on existing Contrast Maximization methods.
The paper tackles motion segmentation in event camera data by iteratively classifying events into background and foreground motions, extending the Contrast Maximization framework. It achieves state-of-the-art accuracy with over 30% improvement on moving object detection benchmarks.
Event cameras provide rich signals that are suitable for motion estimation since they respond to changes in the scene. As any visual changes in the scene produce event data, it is paramount to classify the data into different motions (i.e., motion segmentation), which is useful for various tasks such as object detection and visual servoing. We propose an iterative motion segmentation method, by classifying events into background (e.g., dominant motion hypothesis) and foreground (independent motion residuals), thus extending the Contrast Maximization framework. Experimental results demonstrate that the proposed method successfully classifies event clusters both for public and self-recorded datasets, producing sharp, motion-compensated edge-like images. The proposed method achieves state-of-the-art accuracy on moving object detection benchmarks with an improvement of over 30%, and demonstrates its possibility of applying to more complex and noisy real-world scenes. We hope this work broadens the sensitivity of Contrast Maximization with respect to both motion parameters and input events, thus contributing to theoretical advancements in event-based motion segmentation estimation. https://github.com/aoki-media-lab/event_based_segmentation_vcmax