A probabilistic patch based image representation using Conditional Random Field model for image classification

TL;DR

This paper introduces a novel image classification method that encodes local image patches and their spatial relationships using a Conditional Random Field, improving classification accuracy on standard datasets.

Contribution

The paper presents a new ordered patch-based image representation method using CRF to model spatial relationships, enhancing texture, face, and scene classification.

Findings

Outperforms existing image classification methods on standard datasets

Effectively encodes local properties and spatial relationships

Improves accuracy in texture, face, and scene recognition

Abstract

In this paper we proposed an ordered patch based method using Conditional Random Field (CRF) in order to encode local properties and their spatial relationship in images to address texture classification, face recognition, and scene classification problems. Typical image classification approaches work without considering spatial causality among distinctive properties of an image for image representation in feature space. In this method first, each image is encoded as a sequence of ordered patches, including local properties. Second, the sequence of these ordered patches is modeled as a probabilistic feature vector by CRF to model spatial relationship of these local properties. And finally, image classification is performed on such probabilistic image representation. Experimental results on several standard image datasets indicate that proposed method outperforms some of existing image classification methods.