paper-with-me

홈 › Papers

Continuous Relaxation of MAP Inference: A Nonconvex Perspective

2018-02-21 · CVPR 2018 6 · D. Khuê Lê-Huu, Nikos Paragios

In this paper, we study a nonconvex continuous relaxation of MAP inference in discrete Markov random fields (MRFs). We show that for arbitrary MRFs, this relaxation is tight, and a discrete stationary point of it can be easily reached by a simple block coordinate descent algorithm. In addition, we study the resolution of this relaxation using popular gradient methods, and further propose a more effective solution using a multilinear decomposition framework based on the alternating direction method of multipliers (ADMM). Experiments on many real-world problems demonstrate that the proposed ADMM significantly outperforms other nonconvex relaxation based methods, and compares favorably with state of the art MRF optimization algorithms in different settings.

📄 PDF Abstract BibTeX arXiv:1802.07796

Code (2)

netw0rkf10w/NoRELAX
netw0rkf10w/NoRELAX-cpp

Methods 이 논문이 사용한 방법론

ADMM The alternating direction method of multipliers (ADMM) is an algorithm that solves convex optimization problems by breaking them into smaller pieces, each of which are…

Similar Papers 제목 키워드 기반

Sublabel-Accurate Discretization of Nonconvex Free-Discontinuity Problems

2016-11-21 · ICCV 2017 10 · Thomas Möllenhoff, Daniel Cremers

In this work we show how sublabel-accurate multilabeling approaches can be derived by approximating a classical label-continuous convex relaxation of nonconvex free-discontinuity problems. This insight allows to extend t…

Regularized Frank-Wolfe for Dense CRFs: Generalizing Mean Field and Beyond

2021-10-27 · NeurIPS 2021 12 · Đ. Khuê Lê-Huu, Karteek Alahari

We introduce regularized Frank-Wolfe, a general and effective algorithm for inference and learning of dense conditional random fields (CRFs). The algorithm optimizes a nonconvex continuous relaxation of the CRF inference…

Semantic Segmentation

Lifting the Convex Conjugate in Lagrangian Relaxations: A Tractable Approach for Continuous Markov Random Fields

2021-07-13 · Hartmut Bauermeister, Emanuel Laude, Thomas Möllenhoff, Michael Moeller 외

Dual decomposition approaches in nonconvex optimization may suffer from a duality gap. This poses a challenge when applying them directly to nonconvex problems such as MAP-inference in a Markov random field (MRF) with co…

Stereo Matching

Sparse Signal Reconstruction for Nonlinear Models via Piecewise Rational Optimization

2020-10-29 · Arthur Marmin, Marc Castella, Jean-Christophe Pesquet, Laurent Duval

We propose a method to reconstruct sparse signals degraded by a nonlinear distortion and acquired at a limited sampling rate. Our method formulates the reconstruction problem as a nonconvex minimization of the sum of a d…

Sublabel-Accurate Relaxation of Nonconvex Energies

2015-12-04 · CVPR 2016 6 · Thomas Möllenhoff, Emanuel Laude, Michael Moeller, Jan Lellmann 외

We propose a novel spatially continuous framework for convex relaxations based on functional lifting. Our method can be interpreted as a sublabel-accurate solution to multilabel problems. We show that previously proposed…