Complementary Optic Flow on GPU - Mathematical Image Analysis Group, Saarland University

Complementary Optic Flow on the GPU

A Highly Efficient GPU Implementation for
Variational Optic Flow Based on the Euler-Lagrange Framework

Pascal Gwosdek Henning Zimmer Sven Grewenig
Andrés Bruhn Joachim Weickert

Mathematical Image Analysis Group,
Saarland University, Campus E1.1, Saarbrücken, Germany
{gwosdek, zimmer, grewenig, bruhn, weickert}@mia.uni-saarland.de

Abstract
The Euler-Lagrange (EL) framework is the most widely-used strategy for solving variational optic flow methods. We present the first approach that solves the EL equations of state-of-the-art methods on sequences with 640×480 pixels in near-realtime on GPUs. This performance is achieved by combining two ideas:
(i) We extend the recently proposed Fast Explicit Diffusion (FED) scheme to optic flow, and additionally embed it into a coarse-to-fine strategy. (ii) We parallelise our complete algorithm on a GPU, where a careful optimisation of global memory operations and an efficient use of on-chip memory guarantee a good performance. Applying our approach to the variational ‘Complementary Optic Flow’ method (Zimmer et al. (2009)), we obtain highly accurate flow fields in less than a second. This currently constitutes the fastest method in the top 10 of the widely used Middlebury benchmark.

Welcome to the supplementary material page for our paper A Highly Efficient GPU Implementation for Variational Optic Flow Based on the Euler-Lagrange Framework which has been published at the ECCV Workshop for Computer Vision with GPUs. It constitutes a revised version of Technical Report No. 267.

The aim of these pages is to provide the reader with additional information that go beyond the paper's scope, but which might be useful for a deeper understanding. Please choose one of the following topics: