Infrared and Laser Engineering, Volume. 54, Issue 7, 20250189(2025)
Deep learning empowers generation and presentation of virtual-real fusion scenarios in holographic metaverse: development and prospects (invited)
Fig. 1. (a) Architecture diagram of 3D holographic video communication system; (b) 3D acquisition terminal constructed using structured light camera and light field camera; (c) 3D display terminal constructed using computer-generated holography[5]
Fig. 3. (a) Schematic diagram of supervised monocular depth estimation; (b) Schematic diagram of unsupervised monocular depth estimation
Fig. 4. (a) Two-stage monocular depth estimation; (b) Feature distillation-based monocular depth estimation; (c) Transfer learning-based monocular depth estimation
Fig. 5. (a) Schematic of data-driven hologram generation; (b) Schematic of model-driven hologram generation
Fig. 6. (a) Generation of broadband holograms; (b) Optimization of initial conditions and solution paths for ill-posed inverse problems
Fig. 7. Gradient variation in the depth direction of holographic recon-structions under initial conditions without and with random phase
Get Citation
Copy Citation Text
Zehao HE, Yunhui GAO, Liangcai CAO, Yan ZHANG. Deep learning empowers generation and presentation of virtual-real fusion scenarios in holographic metaverse: development and prospects (invited)[J]. Infrared and Laser Engineering, 2025, 54(7): 20250189
Category: Special issue—Advanced display technology and applications
Received: Mar. 25, 2025
Accepted: --
Published Online: Aug. 29, 2025
The Author Email: Yan ZHANG (yzhang@cnu.edu.cn)