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

10/27/2021
by   D. Khuê Lê-Huu, et al.
0

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 problem using vanilla Frank-Wolfe with approximate updates, which are equivalent to minimizing a regularized energy function. Our proposed method is a generalization of existing algorithms such as mean field or concave-convex procedure. This perspective not only offers a unified analysis of these algorithms, but also allows an easy way of exploring different variants that potentially yield better performance. We illustrate this in our empirical results on standard semantic segmentation datasets, where several instantiations of our regularized Frank-Wolfe outperform mean field inference, both as a standalone component and as an end-to-end trainable layer in a neural network. We also show that dense CRFs, coupled with our new algorithms, produce significant improvements over strong CNN baselines.

READ FULL TEXT

page 1

page 2

page 3

page 4

research
08/22/2016

Efficient Continuous Relaxations for Dense CRF

Dense conditional random fields (CRF) with Gaussian pairwise potentials ...
research
02/12/2023

Mean Field Optimization Problem Regularized by Fisher Information

Recently there is a rising interest in the research of mean field optimi...
research
05/23/2018

Efficient Relaxations for Dense CRFs with Sparse Higher Order Potentials

Dense conditional random fields (CRFs) with Gaussian pairwise potentials...
research
02/01/2022

Dilated Continuous Random Field for Semantic Segmentation

Mean field approximation methodology has laid the foundation of modern C...
research
06/20/2023

Mean-field Analysis of Generalization Errors

We propose a novel framework for exploring weak and L_2 generalization e...
research
12/06/2019

End-to-end Training of CNN-CRF via Differentiable Dual-Decomposition

Modern computer vision (CV) is often based on convolutional neural netwo...
research
07/04/2012

The DLR Hierarchy of Approximate Inference

We propose a hierarchy for approximate inference based on the Dobrushin,...

Please sign up or login with your details

Forgot password? Click here to reset