KOREATECH-CGH Releases Large-Scale RGB-D Hologram Dataset
Researchers present KOREATECH-CGH, a publicly available dataset of 6,000 RGB-D image and complex hologram pairs spanning resolutions from 256×256 to 2048×2048 and extensive depth ranges. They introduce amplitude projection, a post-processing technique that replaces amplitude per depth layer while preserving phase, boosting reconstruction fidelity to 27.01 dB PSNR and 0.87 SSIM. Experiments validate the dataset for ML-CGH training and super-resolution evaluation.
Key Points
- 1Publishes 6,000 RGB-D and complex hologram pairs across 256×256–2048×2048 and extended depth ranges.
- 2Introduces amplitude projection that replaces amplitude per depth layer while preserving phase, improving fidelity.
- 3Enables higher-quality ML-CGH training and evaluation, achieving 27.01 dB PSNR and 0.87 SSIM.
Scoring Rationale
Substantial dataset and fidelity gains drive impact, but scope is niche and findings come from a single preprint source.
Sources
Public references used for this report.
Practice interview problems based on real data
1,625 SQL & Python problems across 15 industry datasets — the exact type of data you work with.
Try 250 free problems
