Relative Attribute Classification with Deep Rank SVM

TL;DR

This paper presents Deep Rank SVM, a deep Siamese network that effectively ranks image pairs based on attribute strength, outperforming previous methods on multiple benchmark datasets.

Contribution

Introduction of Deep Rank SVM, a joint end-to-end deep learning model combining Siamese architecture with rank SVM loss for relative attribute classification.

Findings

Outperforms state-of-the-art on three of four datasets

Effective joint learning of features and ranking function

Demonstrates high accuracy in attribute ranking tasks

Abstract

Relative attributes indicate the strength of a particular attribute between image pairs. We introduce a deep Siamese network with rank SVM loss function, called Deep Rank SVM (DRSVM), in order to decide which one of a pair of images has a stronger presence of a specific attribute. The network is trained in an end-to-end fashion to jointly learn the visual features and the ranking function. We demonstrate the effectiveness of our approach against the state-of-the-art methods on four image benchmark datasets: LFW-10, PubFig, UTZap50K-lexi and UTZap50K-2 datasets. DRSVM surpasses state-of-art in terms of the average accuracy across attributes, on three of the four image benchmark datasets.