“Assessing Learned Models for Phase-only Hologram Compression” by Peng, Zhan, Spjut and Akşit
Conference:
Type(s):
Title:
- Assessing Learned Models for Phase-only Hologram Compression
Session/Category Title:
- Images, Video & Computer Vision
Presenter(s)/Author(s):
Abstract:
We evaluate four models using INR and VAE structures for compressing phase-only holograms. Our findings show that the pretrained VAE struggles with this task, while SIREN achieves 40% compression with high-quality 3D images (PSNR = 34.54 dB), highlighting the effectiveness of INRs and VAE limitations.
References:
[1] Ollin Boer Bohan. 2023. Tiny Autoencoder for Stable Diffusion. https://github.com/madebyollin/taesd. Accessed: June 2025.
[2] Eric R Chan, Marco Monteiro, Petr Kellnhofer, Jiajun Wu, and Gordon Wetzstein. 2021. pi-gan: Periodic implicit generative adversarial networks for 3d-aware image synthesis. In Proceedings of the IEEE/CVF conference on computer vision and pattern recognition. 5799–5809.
[3] Koray Kavaklı, Liang Shi, Hakan Urey, Wojciech Matusik, and Kaan Akşit. 2023. Multi-color Holograms Improve Brightness in Holographic Displays. In SIGGRAPH ASIA 2023 Conference Papers (Sydney, NSW, Australia) (SA ’23). Article 20, 11 pages.
[4] Vincent Sitzmann, Julien Martel, Alexander Bergman, David Lindell, and Gordon Wetzstein. 2020. Implicit neural representations with periodic activation functions. Advances in neural information processing systems 33 (2020), 7462–7473.
[5] Yujie Wang, Praneeth Chakravarthula, Qi Sun, and Baoquan Chen. 2022. Joint neural phase retrieval and compression for energy-and computation-efficient holography on the edge. ACM Transactions on Graphics 41, 4 (2022).
[6] Chuanjun Zheng, Yicheng Zhan, Liang Shi, Ozan Cakmakci, and Kaan Akşit. 2024. Focal Surface Holographic Light Transport using Learned Spatially Adaptive Convolutions. In SIGGRAPH Asia 2024 Technical Communications (SA Technical Communications ’24) (Tokyo, Japan) (SA ’24).
