A massively parallel multi-level approach to a domain decomposition method for the optical flow estimation with varying illumination

TL;DR

This paper introduces a multi-level parallel domain decomposition method for optical flow estimation under varying illumination, combining adaptive regularization with high-performance computing techniques to improve efficiency and accuracy.

Contribution

It presents a novel multi-level parallel framework integrating domain decomposition and adaptive regularization for optical flow with varying illumination.

Findings

Efficiently computes optical flow on high-resolution images.

Preserves edges and fine features in flow estimation.

Validated on classical and real-world image sequences.

Abstract

We consider a variational method to solve the optical flow problem with varying illumination. We apply an adaptive control of the regularization parameter which allows us to preserve the edges and fine features of the computed flow. To reduce the complexity of the estimation for high resolution images and the time of computations, we implement a multi-level parallel approach based on the domain decomposition with the Schwarz overlapping method. The second level of parallelism uses the massively parallel solver MUMPS. We perform some numerical simulations to show the efficiency of our approach and to validate it on classical and real-world image sequences.