Globally Optimal Contrast Maximisation for Event-based Motion Estimation

TL;DR

This paper introduces a globally optimal algorithm for event-based motion estimation using contrast maximisation, overcoming local minima issues of previous iterative methods, with applications in video stabilisation and attitude estimation.

Contribution

It presents a novel branch-and-bound based method with rigorously validated bounding functions for globally optimal contrast maximisation in event streams.

Findings

Successfully solves rotational motion estimation on event data.

Provides concrete examples demonstrating the importance of global optimality.

Achieves exact solutions with reasonable processing time, scalable with GPU acceleration.

Abstract

Contrast maximisation estimates the motion captured in an event stream by maximising the sharpness of the motion compensated event image. To carry out contrast maximisation, many previous works employ iterative optimisation algorithms, such as conjugate gradient, which require good initialisation to avoid converging to bad local minima. To alleviate this weakness, we propose a new globally optimal event-based motion estimation algorithm. Based on branch-and-bound (BnB), our method solves rotational (3DoF) motion estimation on event streams, which supports practical applications such as video stabilisation and attitude estimation. Underpinning our method are novel bounding functions for contrast maximisation, whose theoretical validity is rigorously established. We show concrete examples from public datasets where globally optimal solutions are vital to the success of contrast maximisation. Despite its exact nature, our algorithm is currently able to process a 50,000 event input in 300 seconds (a locally optimal solver takes 30 seconds on the same input), and has the potential to be further speeded-up using GPUs.