“Real-Time Holographic Media System” by Kwon, Cheon and Hong
Conference:
Experience Type(s):
Title:
- Real-Time Holographic Media System
Organizer(s)/Presenter(s):
Interest Areas(s):
- New Technologies
- Research / Education
Description:
This technology is a real-time holography media system that converts 2D media on a computer screen into 3D holograms in real time. Computer-Generated Holography (CGH) uses algorithms to represent depth in a three-dimensional manner. While increasing depth layers enriches information, it significantly raises computational load. Recent GPU advancements enable real-time CGH generation through parallel programming, but these methods are impractical due to high costs and power consumption. In contrast, dedicated hardware like FPGA (Field-Programmable Gate Array) efficiently calculates CGH with low power consumption and compactness. Currently, the best real-time CGH processors generate 8-layer CGH at 30fps using a 4K spatial light modulator (SLM), but they are limited to pre-prepared RGBD (RGB+Depth) data, with FFT processing size and speed as bottlenecks. This new product overcomes these limitations by applying an efficient CGH generation core and high-speed multi-channel FFT using HBM memory. It can convert common 2D videos such as YouTube and Netflix to 3D CGH in real time. Key features include: – An FPGA-based Real-time Holography Processor (RHP) designed with multi-channel HBM memory and efficient algorithms. – A CGH generation hardware architecture using Fresnel propagation equations and layer-based algorithms. – Full-color volumetric holograms with 8-layer depth at 30fps using a 4K SLM. – Low RHP processor latency of 28ms (RGBD to CGH). – Software for RGBD + Sound packet transmission and real-time depth estimation from 2D videos. In this product, we present a real-time holographic media system capable of generating 4K resolution, 8-layer depth holograms at 30 fps using 2D video input without depth information. The proposed system can convert any 2D video into holograms in real time with low power consumption and a compact form factor, making it suitable for various applications such as holographic video streaming, and automotive holographic head-up displays.
References:
[1] Hojung Kim, Yongkyu Kim, Hyunwook Ji, Hyunsik Park, Jungkwuen An, Hoon Song, Yun Tae Kim, Hong-Seok Lee, and Kichul Kim. 2019. A Single-Chip FPGA Holographic Video Processor. IEEE Transactions on Industrial Electronics 66, 3 (2019), 2066–2073.
[2] Wonok Kwon, Sanghoon Cheon, and Keehoon Hong. 2024. Real-time Holographic Media System Utilizing HBM-based Holography Processor. SIGGRAPH Asia 2024 Posters (2024).
[3] Jae-Hyeung Park. 2017. Recent progress in computer-generated holography for three-dimensional scenes. Journal of Information Display 18, 1 (2017), 1–12.
