Latent Feature Detection using LDA in Interior Image

Abstract

Interior image recommendations have been paid attention to with the development of image retrieval technology. Although a classification model is often used to model user-preferred interiors, it is not suitable for explaining why users prefer them. On the other hand, LDA, which analyzes sentence features based on latent features, can provide a better understanding of sentences by estimating the latent features in the sentences. Therefore, in order to understand the structure of preferred interior images by users, we propose a method to estimate latent features in interior images by incorporating spatial features of images and LDA. Specifically, instead of sentences, we use interior images to extract spatial features using BoVW based on SURF features and their histograms, and estimate topics based on the histograms by LDA. We compared the proposed method with subject experiments in interior images with several styles and verified that the proposed method can estimate similar interior topics compared to the subject ’ s subjective view.