VALSE 论文速览 第196期：基于LDR全景图的HDR新视角和几何合成

论文题目：

Pano-NeRF: Synthesizing High Dynamic Range Novel Views with Geometry from Sparse Low Dynamic Range Panoramic Images

作者列表：

陆展 (新加坡南洋理工大学)、郑乾 (浙江大学)、施柏鑫 (北京大学)、蒋旭东 (新加坡南洋理工大学)

B站观看网址：

https://www.bilibili.com/video/BV1WJxKe1Eqi/?spm_id_from=333.999.0.0

复制链接到浏览器打开或点击阅读原文即可跳转至观看页面。

论文摘要：

Panoramic imaging research on geometry recovery and High Dynamic Range (HDR) reconstruction becomes a trend with the development of Extended Reality (XR). Neural Radiance Fields (NeRF) provide a promising scene representation for both tasks without requiring extensive prior data. However, in the case of inputting sparse Low Dynamic Range (LDR) panoramic images, NeRF often degrades with under-constrained geometry and is unable to reconstruct HDR radiance from LDR inputs. We will introduce the observation that the radiance from each pixel in panoramic images can be modelled as both a signal to convey scene lighting information and a light source to illuminate other pixels. Hence, this talk will present the irradiance fields from sparse LDR panoramic images, and we will also discuss how irradiance fields benefit the panoramic HDR reconstruction and geometry.

参考文献：

[1] Zhan Lu, Qian Zheng, Boxin Shi, Xudong Jiang. Pano-NeRF: Synthesizing High Dynamic Range Novel Views with Geometry from Sparse Low Dynamic Range Panoramic Images. The 38th Annual AAAI Conference on Artificial Intelligence (AAAI), Vancouver, Canada, 2024.

论文链接：

[https://arxiv.org/abs/2312.15942]

代码链接：

[https://github.com/Lu-Zhan/Pano-NeRF]

视频讲者简介：

Zhan Lu is a Ph.D. student at the Department of Computer Science and Engineering of Nanyang Technological University. He received the Master degree in Electrical and Computer Engineering from Nanyang Technological University in 2020.

特别鸣谢本次论文速览主要组织者：

月度轮值AC：王一帆 (大连理工大学)