Human-In-The-Loop Person Re-Identification

TL;DR

This paper introduces a human-in-the-loop person re-identification model that learns incrementally from human feedback, eliminating the need for pre-labelled data and enabling scalable, real-time re-id in large camera networks.

Contribution

The paper proposes HVIL, a novel incremental learning approach for re-id that adapts on-the-fly without pre-labelled data, and RMEL, an ensemble method for when human feedback stops.

Findings

HVIL improves re-id ranking accuracy with human feedback.

RMEL effectively combines multiple HVIL models into a robust ensemble.

The approach scales to large gallery sizes and new camera pairs.

Abstract

Current person re-identification (re-id) methods assume that (1) pre-labelled training data are available for every camera pair, (2) the gallery size for re-identification is moderate. Both assumptions scale poorly to real-world applications when camera network size increases and gallery size becomes large. Human verification of automatic model ranked re-id results becomes inevitable. In this work, a novel human-in-the-loop re-id model based on Human Verification Incremental Learning (HVIL) is formulated which does not require any pre-labelled training data to learn a model, therefore readily scalable to new camera pairs. This HVIL model learns cumulatively from human feedback to provide instant improvement to re-id ranking of each probe on-the-fly enabling the model scalable to large gallery sizes. We further formulate a Regularised Metric Ensemble Learning (RMEL) model to combine a series of incrementally learned HVIL models into a single ensemble model to be used when human feedback becomes unavailable.