穿透明暗：基于传感器物理建模的单曝光图像与事件去模糊 HDR 神经辐射场
Seeing through Light and Darkness: Sensor-Physics Grounded Deblurring HDR NeRF from Single-Exposure Images and Events
把相机响应函数（CRF）与事件传感器响应模型显式嵌入 NeRF 优化，用物理传感器模型对齐渲染值与 LDR 观测，这一思路可直接迁移到事件相机、卷帘快门、RAW 域等存在传感器非线性失配的重建与逆渲染任务。
原文依据 A 2D pixel-wise RGB CRF model is introduced to align the NeRF rendered pixel values with the sensor-recorded LDR pixel values
摘要译文
由模型翻译 · 术语统一 · 请对照原文核对关键结论从低动态范围（LDR）模糊图像进行新视角合成在真实场景中十分常见，但这类方法在极端光照条件下难以恢复高动态范围（HDR）且清晰的三维表示。尽管现有方法借助事件数据来应对该问题，但它们忽略了相机输出与物理世界辐射亮度之间的传感器物理失配，导致 HDR 与去模糊效果欠佳。为解决该问题，我们提出一个统一的、以传感器物理为基础的 NeRF 框架，从单曝光模糊 LDR 图像及对应事件中实现清晰 HDR 新视角合成。我们利用 NeRF 直接在 HDR 域表示三维场景的真实辐射亮度，并像物理世界那样对击中传感器像素的原始 HDR 场景光线进行建模。我们引入一个二维逐像素 RGB 相机响应函数（CRF）模型，使 NeRF 渲染的像素值与输入图像中传感器记录的 LDR 像素值对齐。我们还设计了一个新的事件 CRF 模型，以弥合物理场景动态与事件传感器输出之间的差距。这两个模型与 NeRF 网络联合优化，利用事件中的空间与时序动态信息来增强清晰 HDR 三维表示的学习。在自采集数据集与公开数据集上的实验表明，我们的方法在单曝光模糊 LDR 图像及对应事件输入下取得了最优的 HDR 与去模糊新视角合成结果。我们的代码与数据集已公开于 https://github.com/iCVTEAM/See-NeRF。
英文原文摘要 1519 字符
Novel view synthesis from low dynamic range (LDR) blurry images, which are common in the wild, struggles to recover high dynamic range (HDR) and sharp 3D representations in extreme lighting conditions. Although existing methods employ event data to address this issue, they ignore the sensor-physics mismatches between the camera output and physical world radiance, resulting in suboptimal HDR and deblurring results. To cope with this problem, we propose a unified sensor-physics grounded NeRF framework for sharp HDR novel view synthesis from single-exposure blurry LDR images and corresponding events. We employ NeRF to directly represent the actual radiance of the 3D scene in the HDR domain and model raw HDR scene rays hitting the sensor pixels as in the physical world. A 2D pixel-wise RGB CRF model is introduced to align the NeRF rendered pixel values with the sensor-recorded LDR pixel values of the input images. A novel event CRF model is also designed to bridge the gap between physical scene dynamics and event sensor output. The two models are jointly optimized with the NeRF network, leveraging the spatial and temporal dynamic information in events to enhance the sharp HDR 3D representation learning. Experiments on the collected and public datasets demonstrate that our method achieves state-of-the-art HDR and deblurring novel view synthesis results with single-exposure blurry LDR images and corresponding events. Our code and datasets are publicly available at https://github.com/iCVTEAM/See-NeRF.