arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 2305.01643cs.CV
收藏

用于新视角合成的神经LiDAR场

Neural LiDAR Fields for Novel View Synthesis

ETH Zurich(苏黎世联邦理工学院);NVIDIA(英伟达);University of Toronto(多伦多大学);Vector Institute(向量研究所)

Shengyu Huang, Zan Gojcic, Zian Wang, Francis Williams, Yoni Kasten, Sanja Fidler, Konrad Schindler, Or Litany

AI总结:

提出NFL,将神经场渲染与物理LiDAR传感模型结合以合成新视角扫描,并在真实与合成数据上验证其优于重建模拟及NeRF风格方法。

摘要:

我们提出了Neural Fields for LiDAR(NFL),这是一种从LiDAR测量数据优化神经场场景表示的方法,目标是从新颖视角合成逼真的LiDAR扫描。NFL将神经场的渲染能力与详细、具有物理依据的LiDAR传感过程模型相结合,从而能够准确再现关键传感器行为,如光束发散、二次回波和射线丢失。我们在合成和真实LiDAR扫描上评估了NFL,结果表明,在LiDAR新视角合成任务中,它优于显式的“先重建后模拟”方法以及其他NeRF风格方法。此外,我们证明,合成视图增强的真实感缩小了与真实扫描之间的域差距,并转化为更好的配准和语义分割性能。

英文摘要:

We present Neural Fields for LiDAR (NFL), a method to optimise a neural field scene representation from LiDAR measurements, with the goal of synthesizing realistic LiDAR scans from novel viewpoints. NFL combines the rendering power of neural fields with a detailed, physically motivated model of the LiDAR sensing process, thus enabling it to accurately reproduce key sensor behaviors like beam divergence, secondary returns, and ray dropping. We evaluate NFL on synthetic and real LiDAR scans and show that it outperforms explicit reconstruct-then-simulate methods as well as other NeRF-style methods on LiDAR novel view synthesis task. Moreover, we show that the improved realism of the synthesized views narrows the domain gap to real scans and translates to better registration and semantic segmentation performance.

机构、总结、中文摘要均由AI分析而来,可能存在错误,请以论文正文为准。

↑