MovieNet-PS: A Large-Scale Person Search Dataset in the Wild

TL;DR

This paper introduces GLCNet, a unified global-local context network that leverages scene and group context to improve person search accuracy, validated on multiple datasets including a new large-scale dataset from MovieNet.

Contribution

The paper presents a novel unified network that simultaneously exploits scene and group context for person search, enhancing feature discriminability and performance.

Findings

Consistent improvement over state-of-the-art on CUHK-SYSU, PRW, and MovieNet datasets.

Effective utilization of scene and group context enhances person re-identification.

Public release of dataset, code, and models facilitates future research.

Abstract

Person search aims to jointly localize and identify a query person from natural, uncropped images, which has been actively studied over the past few years. In this paper, we delve into the rich context information globally and locally surrounding the target person, which we refer to as scene and group context, respectively. Unlike previous works that treat the two types of context individually, we exploit them in a unified global-local context network (GLCNet) with the intuitive aim of feature enhancement. Specifically, re-ID embeddings and context features are simultaneously learned in a multi-stage fashion, ultimately leading to enhanced, discriminative features for person search. We conduct the experiments on two person search benchmarks (i.e., CUHK-SYSU and PRW) as well as extend our approach to a more challenging setting (i.e., character search on MovieNet). Extensive experimental results demonstrate the consistent improvement of the proposed GLCNet over the state-of-the-art methods on all three datasets. Our source codes, pre-trained models, and the new dataset are publicly available at: https://github.com/ZhengPeng7/GLCNet.