Version 1
: Received: 27 September 2024 / Approved: 27 September 2024 / Online: 30 September 2024 (07:29:11 CEST)
How to cite:
Kim, Y.; Kim, S.; Min, S.; Han, Y.; Lee, O.; Kim, W. A Dual-Module System for Copyright-Free Image Recommendation and Infringement Detection in Educational Materials. Preprints2024, 2024092285. https://doi.org/10.20944/preprints202409.2285.v1
Kim, Y.; Kim, S.; Min, S.; Han, Y.; Lee, O.; Kim, W. A Dual-Module System for Copyright-Free Image Recommendation and Infringement Detection in Educational Materials. Preprints 2024, 2024092285. https://doi.org/10.20944/preprints202409.2285.v1
Kim, Y.; Kim, S.; Min, S.; Han, Y.; Lee, O.; Kim, W. A Dual-Module System for Copyright-Free Image Recommendation and Infringement Detection in Educational Materials. Preprints2024, 2024092285. https://doi.org/10.20944/preprints202409.2285.v1
APA Style
Kim, Y., Kim, S., Min, S., Han, Y., Lee, O., & Kim, W. (2024). A Dual-Module System for Copyright-Free Image Recommendation and Infringement Detection in Educational Materials. Preprints. https://doi.org/10.20944/preprints202409.2285.v1
Chicago/Turabian Style
Kim, Y., Ohyoung Lee and Wongyum Kim. 2024 "A Dual-Module System for Copyright-Free Image Recommendation and Infringement Detection in Educational Materials" Preprints. https://doi.org/10.20944/preprints202409.2285.v1
Abstract
Images are widely used in educational materials due to their ability to effectively convey complex concepts. However, unauthorized image use often leads to legal issues related to copyright infringement. To address this issue, we introduce a dual-module system designed specifically for educators. The first module, a copyright infringement detection system, leverages deep learning techniques to verify the copyright status of images. It employs a Convolutional Variational Autoencoder (CVAE) model to extract meaningful features from copyrighted images and compares them against user-provided images. If infringement is suspected, the second module, an image retrieval system, recommends alternative copyright-free images using a Vision Transformer (ViT)-based hashing model. Evaluation on benchmark datasets demonstrates the system's effectiveness, achieving a mean Average Precision (mAP) of 0.812 on the Flickr25k dataset. Furthermore, a user study with 65 teachers indicates high satisfaction levels, particularly in addressing copyright concerns and ease of use. Our system significantly aids educators in creating educational materials compliant with copyright regulations.
Computer Science and Mathematics, Artificial Intelligence and Machine Learning
Copyright:
This is an open access article distributed under the Creative Commons Attribution License which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.