Comparator Networks

TL;DR

This paper introduces Deep Comparator Networks (DCN), a novel neural architecture for set-based face verification that directly learns to compare sets of images by attending to local regions, outperforming previous methods.

Contribution

The paper presents a new neural network architecture that directly compares sets of images for verification, incorporating local region attention and a novel hard sample mining regime.

Findings

DCN outperforms previous state-of-the-art on Janus face recognition benchmarks.

The model effectively attends to discriminative local regions for set comparison.

Hard sample mining improves the model's robustness and accuracy.

Abstract

The objective of this work is set-based verification, e.g. to decide if two sets of images of a face are of the same person or not. The traditional approach to this problem is to learn to generate a feature vector per image, aggregate them into one vector to represent the set, and then compute the cosine similarity between sets. Instead, we design a neural network architecture that can directly learn set-wise verification. Our contributions are: (i) We propose a Deep Comparator Network (DCN) that can ingest a pair of sets (each may contain a variable number of images) as inputs, and compute a similarity between the pair--this involves attending to multiple discriminative local regions (landmarks), and comparing local descriptors between pairs of faces; (ii) To encourage high-quality representations for each set, internal competition is introduced for recalibration based on the landmark score; (iii) Inspired by image retrieval, a novel hard sample mining regime is proposed to control the sampling process, such that the DCN is complementary to the standard image classification models. Evaluations on the IARPA Janus face recognition benchmarks show that the comparator networks outperform the previous state-of-the-art results by a large margin.