Manga Text Detection with Manga-Specific Data Augmentation and Its Applications on Emotion Analysis

Yi Ting Yang, Wei Ta Chu

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

Abstract

We especially target at detecting text in atypical font styles and in cluttered background for Japanese comics (manga). To enable the detection model to detect atypical text, we augment training data by the proposed manga-specific data augmentation. A generative adversarial network is developed to generate atypical text regions, which are then blended into manga pages to largely increase the volume and diversity of training data. We verify the importance of manga-specific data augmentation. Furthermore, with the help of manga text detection, we fuse global visual features and local text features to enable more accurate emotion analysis.

Original languageEnglish
Title of host publicationMultiMedia Modeling - 29th International Conference, MMM 2023, Proceedings
EditorsDuc-Tien Dang-Nguyen, Cathal Gurrin, Alan F. Smeaton, Martha Larson, Stevan Rudinac, Minh-Son Dao, Christoph Trattner, Phoebe Chen
PublisherSpringer Science and Business Media Deutschland GmbH
Pages29-40
Number of pages12
ISBN (Print)9783031278174
DOIs
Publication statusPublished - 2023
Event29th International Conference on MultiMedia Modeling, MMM 2023 - Bergen, Norway
Duration: 2023 Jan 92023 Jan 12

Publication series

NameLecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)
Volume13834 LNCS
ISSN (Print)0302-9743
ISSN (Electronic)1611-3349

Conference

Conference29th International Conference on MultiMedia Modeling, MMM 2023
Country/TerritoryNorway
CityBergen
Period23-01-0923-01-12

All Science Journal Classification (ASJC) codes

  • Theoretical Computer Science
  • General Computer Science

Fingerprint

Dive into the research topics of 'Manga Text Detection with Manga-Specific Data Augmentation and Its Applications on Emotion Analysis'. Together they form a unique fingerprint.

Cite this