Explainable Person Re-Identification with Attribute-guided Metric Distillation

TL;DR

This paper introduces Attribute-guided Metric Distillation (AMD), a post-hoc explanation method for person re-identification models that clarifies attribute contributions and improves model interpretability and accuracy.

Contribution

The paper presents the first attribute-guided explanation approach for ReID models, including a pluggable interpreter network and novel loss functions for better interpretability and bias reduction.

Findings

Interpreter generates effective attribute-based explanations.

Method improves target model accuracy.

Generalizes well across different datasets.

Abstract

Despite the great progress of person re-identification (ReID) with the adoption of Convolutional Neural Networks, current ReID models are opaque and only outputs a scalar distance between two persons. There are few methods providing users semantically understandable explanations for why two persons are the same one or not. In this paper, we propose a post-hoc method, named Attribute-guided Metric Distillation (AMD), to explain existing ReID models. This is the first method to explore attributes to answer: 1) what and where the attributes make two persons different, and 2) how much each attribute contributes to the difference. In AMD, we design a pluggable interpreter network for target models to generate quantitative contributions of attributes and visualize accurate attention maps of the most discriminative attributes. To achieve this goal, we propose a metric distillation loss by which the interpreter learns to decompose the distance of two persons into components of attributes with knowledge distilled from the target model. Moreover, we propose an attribute prior loss to make the interpreter generate attribute-guided attention maps and to eliminate biases caused by the imbalanced distribution of attributes. This loss can guide the interpreter to focus on the exclusive and discriminative attributes rather than the large-area but common attributes of two persons. Comprehensive experiments show that the interpreter can generate effective and intuitive explanations for varied models and generalize well under cross-domain settings. As a by-product, the accuracy of target models can be further improved with our interpreter.