Transfer Deep Learning for Reconfigurable Snapshot HDR Imaging Using Coded Masks,Computer Graphics Forum

当前位置： X-MOL 学术 › Comput. Graph. Forum › 论文详情

Our official English website, www.x-mol.net, welcomes your feedback! (Note: you will need to create a separate account there.)

Transfer Deep Learning for Reconfigurable Snapshot HDR Imaging Using Coded Masks
Computer Graphics Forum ( IF 2.7 ) Pub Date : 2021-03-11 , DOI: 10.1111/cgf.14205
Masheal Alghamdi ₁ , Qiang Fu ₁ , Ali Thabet ₁ , Wolfgang Heidrich ₁

Affiliation

High dynamic range (HDR) image acquisition from a single image capture, also known as snapshot HDR imaging, is challenging because the bit depths of camera sensors are far from sufficient to cover the full dynamic range of the scene. Existing HDR techniques focus either on algorithmic reconstruction or hardware modification to extend the dynamic range. In this paper we propose a joint design for snapshot HDR imaging by devising a spatially varying modulation mask in the hardware and building a deep learning algorithm to reconstruct the HDR image. We leverage transfer learning to overcome the lack of sufficiently large HDR datasets available. We show how transferring from a different large-scale task (image classification on ImageNet) leads to considerable improvements in HDR reconstruction. We achieve a reconfigurable HDR camera design that does not require custom sensors, and instead can be reconfigured between HDR and conventional mode with very simple calibration steps. We demonstrate that the proposed hardwareâ€“software so lution offers a flexible yet robust way to modulate per-pixel exposures, and the network requires little knowledge of the hardware to faithfully reconstruct the HDR image. Comparison results show that our method outperforms the state of the art in terms of visual perception quality.

中文翻译：

使用编码掩码为可重构快照 HDR 成像迁移深度学习

从单个图像捕获中获取高动态范围 (HDR) 图像（也称为快照 HDR 成像）具有挑战性，因为相机传感器的位深度远不足以覆盖场景的整个动态范围。现有的 HDR 技术侧重于算法重建或硬件修改以扩展动态范围。在本文中，我们通过在硬件中设计空间变化的调制掩码并构建深度学习算法来重建 HDR 图像，提出了快照 HDR 成像的联合设计。我们利用迁移学习来克服缺乏足够大的可用 HDR 数据集的问题。我们展示了从不同的大规模任务（ImageNet 上的图像分类）转移如何导致 HDR 重建的显着改进。我们实现了不需要定制传感器的可重新配置的 HDR 相机设计，而是可以通过非常简单的校准步骤在 HDR 和传统模式之间重新配置。我们证明了所提出的硬件-软件解决方案提供了一种灵活而稳健的方式来调制每像素曝光，并且网络需要很少的硬件知识来忠实地重建 HDR 图像。比较结果表明，我们的方法在视觉感知质量方面优于现有技术。并且网络需要很少的硬件知识来忠实地重建 HDR 图像。比较结果表明，我们的方法在视觉感知质量方面优于现有技术。并且网络需要很少的硬件知识来忠实地重建 HDR 图像。比较结果表明，我们的方法在视觉感知质量方面优于现有技术。

更新日期：2021-03-11

点击分享查看原文

点击收藏

阅读更多本刊最新论文本刊介绍/投稿指南11