GPU-Based Multigrid: Real-Time Performance in High Resolution Nonlinear Image Processing

Grossauer, Harald; Thoman, Peter

doi:10.1007/978-3-540-79547-6_14

Cited by 21 publications

(24 citation statements)

References 14 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…On serial hardware, multi-grid solvers based on GaussSeidel have been proposed in [9]. A GPU implementation of the formulation in [9] has been proposed using Jacobi solvers [10]. Compared to [10], our implementation handles large displacements through dense descriptor matching.…”

Section: Related Workmentioning

confidence: 99%

See 1 more Smart Citation

Dense Point Trajectories by GPU-Accelerated Large Displacement Optical Flow

Sundaram

Brox

Keutzer

2010

Lecture Notes in Computer Science

370

349

View full text Add to dashboard Cite

Dense and accurate motion tracking is an important requirement for many video feature extraction algorithms. In this paper we provide a method for computing point trajectories based on a fast parallel implementation of a recent optical flow algorithm that tolerates fast motion. The parallel implementation of large displacement optical flow runs about 78× faster than the serial C++ version. This makes it practical to use in a variety of applications, among them point tracking. In the course of obtaining the fast implementation, we also prove that the fixed point matrix obtained in the optical flow technique is positive semi-definite. We compare the point tracking to the most commonly used motion tracker -the KLT tracker -on a number of sequences with ground truth motion. Our resulting technique tracks up to three orders of magnitude more points and is 46% more accurate than the KLT tracker. It also provides a tracking density of 48% and has an occlusion error of 3% compared to a density of 0.1% and occlusion error of 8% for the KLT tracker. Compared to the Particle Video tracker, we achieve 66% better accuracy while retaining the ability to handle large displacements while running an order of magnitude faster.

show abstract

Section: Related Workmentioning

confidence: 99%

“…A GPU implementation of the formulation in [9] has been proposed using Jacobi solvers [10]. Compared to [10], our implementation handles large displacements through dense descriptor matching. Such extensions enable us to handle fast motion well [11], [4].…”

Section: Related Workmentioning

confidence: 99%

Dense Point Trajectories by GPU-Accelerated Large Displacement Optical Flow

Sundaram

Brox

Keutzer

2010

Lecture Notes in Computer Science

370

349

View full text Add to dashboard Cite

show abstract

“…It is also the most expensive GPU kernel in our framework: Due to its low arithmetic complexity, it is strictly memory bound and requires significant amounts of data. For the smoothness term, we reduce the memory complexity by exploiting the symmetry of the non-diagonal matrix A from (14), which comes down to store the four upper off-diagonals. The remaining entries can be computed in shared memory.…”

Section: Implementation On the Gpumentioning

confidence: 99%

“…Modern multigrid methods are well-known for their good performance on CPUs [12,13], but still do not achieve even near-realtime performance on larger image sequences. Multigrid methods on GPUs do achieve realtime performance, but due to their complicated implementation, they were only realised for basic models so far [14]. Another class of efficient algorithms that can easily be parallelised for GPUs and additionally support modern models are primal-dual approaches; see e.g.…”

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

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

Gwosdek

Zimmer

Grewenig

et al. 2012

Lecture Notes in Computer Science

View full text Add to dashboard Cite

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.

show abstract

Parallel Unsmoothed Aggregation Algebraic Multigrid Algorithms on GPUs

Brannick

Chen

et al. 2013

Numerical Solution of Partial Differential Equations: Theory, Algorithms, and Their Applications

View full text Add to dashboard Cite

We design and implement a parallel algebraic multigrid method for isotropic graph Laplacian problems on multicore Graphical Processing Units (GPUs). The proposed AMG method is based on the aggregation framework. The setup phase of the algorithm uses a parallel maximal independent set algorithm in forming aggregates and the resulting coarse level hierarchy is then used in a K-cycle iteration solve phase with a 1 -Jacobi smoother. Numerical tests of a parallel implementation of the method for graphics processors are presented to demonstrate its effectiveness.

show abstract

GPU-Based Multigrid: Real-Time Performance in High Resolution Nonlinear Image Processing

Cited by 21 publications

References 14 publications

Dense Point Trajectories by GPU-Accelerated Large Displacement Optical Flow

Dense Point Trajectories by GPU-Accelerated Large Displacement Optical Flow

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

Parallel Unsmoothed Aggregation Algebraic Multigrid Algorithms on GPUs

Contact Info

Product

Resources

About