Graph-structure based multi-label prediction and classification for unsupervised person re-identification

This paper addresses unsupervised person re-identification (Re-ID) using multi-label prediction and classification based on graph-structural insight. Our method extracts features from person images and produces a graph that consists of the features and a pairwise similarity of them as nodes and edge...

Full description

Saved in:
Bibliographic Details
Published in:Applied intelligence (Dordrecht, Netherlands) Netherlands), 2022-09, Vol.52 (12), p.14281-14293
Main Authors: Yu, Jongmin, Oh, Hyeontaek
Format: Article
Language:English
Subjects:
Artificial Intelligence
Classification
Computer Science
Datasets
Deep learning
Feature extraction
Graph theory
Image classification
Labeling
Labels
Machines
Manufacturing
Mechanical Engineering
Methods
Neural networks
Processes
Similarity
Source code
Citations: Items that this one cites
Items that cite this one
Online Access:Get full text
Tags: Add Tag
No Tags, Be the first to tag this record!
Description
Summary:This paper addresses unsupervised person re-identification (Re-ID) using multi-label prediction and classification based on graph-structural insight. Our method extracts features from person images and produces a graph that consists of the features and a pairwise similarity of them as nodes and edges, respectively. Based on the graph, the proposed graph structure based multi-label prediction (GSMLP) method predicts multi-class labels by considering the pairwise similarity and the adjacency node distribution of each node. The multi-class labels created by GSMLP are applied to the proposed balanced multi-label classification (BMLC) loss. BMLC integrates a hard-sample mining scheme and a multi-label classification. The proposed GSMLP and BMLC boost the performance of unsupervised person Re-ID without any pre-labelled dataset. Experimental results justify the superiority of the proposed method in unsupervised person Re-ID by producing state-of-the-art performance. The source code for this paper is publicly available on https://github.com/andreYoo/GSMLP-BMLC.git .
ISSN:0924-669X
1573-7497
DOI:10.1007/s10489-022-03163-6