Person Re-Identification in Movies/Dramas

Dong Hyuck Im, Yong Seok Seo, Hyeonwoo Kim, Eenjun Hwang, Jihyun Park

Research output: Chapter in Book/Report/Conference proceedingConference contribution

Abstract

In this paper, we present a system for person re-identification in movies and dramas. Our person re-identification method simultaneously learns differences and commonalities between two pairs of images. Features are extracted using three images, and the relationship between the extracted features is constructed into two pairs of feature maps to learn. Our method significantly outperforms on a large data set (CUHK03), and our own re-identification dataset generated from video contents such as movies and dramas. We also show that the proposed model learned on our own data set can be applied to actor re-identification in video content.

Original languageEnglish
Title of host publicationICTC 2020 - 11th International Conference on ICT Convergence
Subtitle of host publicationData, Network, and AI in the Age of Untact
PublisherIEEE Computer Society
Pages1596-1598
Number of pages3
ISBN (Electronic)9781728167589
DOIs
Publication statusPublished - 2020 Oct 21
Event11th International Conference on Information and Communication Technology Convergence, ICTC 2020 - Jeju Island, Korea, Republic of
Duration: 2020 Oct 212020 Oct 23

Publication series

NameInternational Conference on ICT Convergence
Volume2020-October
ISSN (Print)2162-1233
ISSN (Electronic)2162-1241

Conference

Conference11th International Conference on Information and Communication Technology Convergence, ICTC 2020
Country/TerritoryKorea, Republic of
CityJeju Island
Period20/10/2120/10/23

Keywords

  • actor re-identification
  • deep learning
  • person re-identification

ASJC Scopus subject areas

  • Information Systems
  • Computer Networks and Communications

Cite this