Efficient Structured Prediction with Latent Variables for General Graphical Models

TL;DR

This paper introduces a unified framework for structured prediction with latent variables, employing a local entropy approximation and an efficient message passing algorithm, demonstrating superior performance in image segmentation and scene understanding.

Contribution

It presents a novel unified approach that generalizes existing models and guarantees convergence, improving results in computer vision tasks.

Findings

Outperforms latent structured SVMs and HCRFs in experiments

Provides a convergent message passing algorithm

Effective in image segmentation and 3D scene understanding

Abstract

In this paper we propose a unified framework for structured prediction with latent variables which includes hidden conditional random fields and latent structured support vector machines as special cases. We describe a local entropy approximation for this general formulation using duality, and derive an efficient message passing algorithm that is guaranteed to converge. We demonstrate its effectiveness in the tasks of image segmentation as well as 3D indoor scene understanding from single images, showing that our approach is superior to latent structured support vector machines and hidden conditional random fields.