@inproceedings{6392,
  keywords = {Comic recognition, convolutional neural network, Convolutional neural networks, cultural heritage, Cultural heritages, Deep learning, Fast region-based convolutional neural network, Faster R-CNN, Historic preservation, Indigenous intangible heritage, Language processing, Learning systems, NLP, Natural language processing, Natural language processing systems, Natural languages, Region-based},
  author = {R. Sharma and V. Kukreja},
  title = {Deep learning-Based Comic Recognition and Analysis for the Preservation of Indigenous Intangible Heritage},
  abstract = {Indigenous intangible heritage is a valuable part of human culture, passed down through generations. This research explores the use of deep learning techniques to preserve and understand this heritage through comics. By detecting and analyzing important elements like speech bubbles and characters, we aim to uncover the cultural significance of these comics in preserving indigenous heritage. Our evaluation of a deep learning pipeline, using Faster Region-based Convolutional Neural Network (R-CNN) and Bidirectional Encoder Representations from Transformers (BERT) models, demonstrates its accuracy and effectiveness in interpreting indigenous intangible heritage through comics. This research sheds light on the potential of deep learning for preserving and promoting indigenous culture, providing valuable insights for researchers and cultural heritage professionals.},
  pages = {136-140},
  url = {https://www.scopus.com/inward/record.uri?eid=2-s2.0-85190548591&doi=10.1109%2fICSH57060.2023.10482818&partnerID=40&md5=8d63b01aa7c90fc15c23505fd998a70d},
  doi = {10.1109/ICSH57060.2023.10482818},
}