문화재 (pagoda) 복원을 위한 GAN 알고리즘 연구
A Study on GAN Algorithm for Restoration of Cultural Property (pagoda)
0
조회
0
다운로드
4
인용
0
참고문헌
키워드
초록
오늘날 문화재 복원은, 기존 자료와 전문가에 의존하는 것과 더불어 최신 IT 기술을 적용하여복원하고 있다. 하지만 새로운 자료가 나와서 기존 복원이 틀리게 되는 경우, 복원하는데 너무 오랜 시간이 걸릴 때도 있다. 그리고 예상과 다른 결과가 나올 가능성도 있다. 이에 우리는 중요 문화재의 복원을 인공지능을 이용하여 빠르게 복원을 해 보고자 한다. 최근에 Generative Adversarial Networks(GANs) 알고리즘에서 DcGAN[2] 알고리즘이 나오면서 이미지 생성, 복원 분야가 지속해서 발전하고 있다. 이에 본 연구에서는 다양한 GAN 알고리즘을 문화재 복원에 GAN 알고리즘을적용해 보았다. DcGAN과 StyleGAN을 적용하였으며, 유의미한 결과를 얻었다. GAN 알고리즘 중DCGAN과 Style GAN 알고리즘을 실험한 결과 DCGAN 알고리즘은 학습이 진행되었으며, 낮은 해상도로 탑 이미지가 생성되는 것을 확인했다. 그리고 Style GAN 알고리즘에서도 역시 학습이 진행 되었으며, 탑 이미지가 생성되었다. 결론적으로 GAN 알고리즘을 사용하여 높은 해상도의 탑이미지를 구할 수 있게 되었다.
Abstract
Today, the restoration of cultural properties is done by applying the latest IT technology from relying on existing data and experts. However, there are cases where new data are released and the original restoration is incorrect. Also, sometimes it takes too long to restore. And there is a possibility that the results will be different than expected. Therefore, we aim to quickly restore cultural properties using DeepLearning. Recently, so the algorithm DcGAN made in GANs algorithm, and image creation, restoring sectors are constantly evolving. We try to find the optimal GAN algorithm for the restoration of cultural properties among various GAN algorithms. Because the GAN algorithm is used in various fields. In the field of restoring cultural properties, it will show that it can be applied in practice by obtaining meaningful results. As a result of experimenting with the DCGAN and Style GAN algorithms among the GAN algorithms, it was confirmed that the DCGAN algorithm generates a top image with a low resolution.
참고문헌 (21)
- Alec Radford, "Unsupervised Representation Learning with Deep Convolutional Generative Adversarial Networks", ICLR 2016 (2016)
- Tero Karras, "A Style-Based Generator Architecture for Generative Adversarial Networks", Proceedings of the IEEE, 2019, 4401~4410
- David Berthelot, "BEGAN:Boundary Equilibrium Generative Adversarial Networks", arXiv:1703.10717, ~ (2017)
- Tero Karras, "Progressive Growing of GANs for Improved Quality, Stability, and Variation", arXiv:1710.10196, ~ (2018)
- "AIHUB"[URL]
- Taesung Park, "Semantic Image Synthesis with Spatially-Adaptive Normalization", arXiv:1903.07291, ~ (2019)
- Andrew Brock, "Large Scale GAN Training for High Fidelity Natural Image Synthesis", arXiv:1809.11096, ~ (2018)
- Yunjey Choi, "StarGAN : Unified Generative Adversarial Networks for Multi-Domain Image-to-Image Translation", IEEE Conference on Computer Vision and Pattern Recognition (CVPR), pp.8789~8797 (2018)
- Z. Zhang, "Age progression", regression by conditional adversarial autoencoder, 2017, ~
- Mehdi Mirza, "Conditional Generative Adversarial Nets", Computer Science (2014)
- Yunjey Choi, "StarGAN v2:Diverse Image Synthesis for Multiple Domains", Computer Science 2020 IEEE, 2020, ~
- Junbo Zhao, "Energy-based Generative Adversarial Network", arXiv preprint arXiv:1609.03126, ~ (2016)
- Soheil Kolouri, "Generalized Sliced Wasserstein Distances", Advances in Neural Information Processing Systems, 32 (2019)
- X. Chen, "InfoGAN : interpretable representation learning by information maximizing generative adversarial nets", CoRR, abs, 2016
- V. Dumoulin, "A learned representation for artistic style", CoRR, abs, 2016
- L. A. Gatys, "Image style transferusing convolutional neural networks", Proc. CVPR (2016)
- D. P. Kingma, "Auto-encoding variational bayes", ICLR (2014)
- M. Lucic, "Are GANs created equal? a large-scale study", CoRR, abs, 2017
- T. Miyato, "cGANs with projection discriminator", CoRR, abs, 2018 (1802)
- T. Miyato, "normalization for generative adversarial networks", CoRR, abs, 2018 (1802)
- M. Marchesi, "Megapixel size image creation using generative adversarial networks", CoRR, abs, 2017