arXivDaily arXiv每日学术速递 周一至周五更新
arXiv周末暂无论文更新,休息一下吧,周末愉快~~
arXiv 2310.06743cs.LGcs.AI
收藏

基于球谐函数与正弦表示网络的地理位置编码

Geographic Location Encoding with Spherical Harmonics and Sinusoidal Representation Networks

Wageningen University(瓦赫宁根大学);Microsoft Research New England(微软研究院新英格兰分部);Harvard Data Science Initiative(哈佛大学数据科学倡议);Center for Research on Computation and Society(计算与社会研究中心);University of Colorado, Boulder(科罗拉多大学博尔德分校);École Polytechnique Fédérale de Lausanne (EPFL)(洛桑联邦理工学院)

Marc Rußwurm, Konstantin Klemmer, Esther Rolf, Robin Zbinden, Devis Tuia

AI总结:

针对现有全球地理坐标嵌入存在伪影、架构设计关注不足的问题,提出结合球谐基函数与SirenNets的新型位置编码器,实验验证二者结合可在多任务中达到SOTA性能。

摘要:

学习地理空间的表示对于任何整合地理定位数据的机器学习模型都至关重要,应用领域涵盖遥感、生态学或流行病学等。近期研究采用基于双傅里叶球(Double Fourier Sphere, DFS)特征的正弦和余弦投影来嵌入坐标。这类嵌入即使处理全球数据也假定数据域为矩形,可能产生伪影,在两极地区尤为明显。同时,学界对与这些函数嵌入结合的神经网络架构的精确设计关注较少。本研究针对全球分布的地理数据提出了一种新型位置编码器,它将原生定义在球面上的球谐基函数,与可被解释为可学习双傅里叶球嵌入的正弦表示网络(SirenNets)相结合。我们在多个基准数据集和合成评估数据集上,对位置嵌入方法与神经网络架构进行了系统性评估。与以往需要位置编码和神经网络结合才能学习有效表示的方法不同,我们的研究表明,球谐函数和正弦表示网络各自单独使用时就具备竞争力,而二者结合后能在各项任务中取得当前最优性能。模型代码与实验结果可在https://github.com/marccoru/locationencoder获取。

英文摘要:

Learning representations of geographical space is vital for any machine learning model that integrates geolocated data, spanning application domains such as remote sensing, ecology, or epidemiology. Recent work embeds coordinates using sine and cosine projections based on Double Fourier Sphere (DFS) features. These embeddings assume a rectangular data domain even on global data, which can lead to artifacts, especially at the poles. At the same time, little attention has been paid to the exact design of the neural network architectures with which these functional embeddings are combined. This work proposes a novel location encoder for globally distributed geographic data that combines spherical harmonic basis functions, natively defined on spherical surfaces, with sinusoidal representation networks (SirenNets) that can be interpreted as learned Double Fourier Sphere embedding. We systematically evaluate positional embeddings and neural network architectures across various benchmarks and synthetic evaluation datasets. In contrast to previous approaches that require the combination of both positional encoding and neural networks to learn meaningful representations, we show that both spherical harmonics and sinusoidal representation networks are competitive on their own but set state-of-the-art performances across tasks when combined. The model code and experiments are available at https://github.com/marccoru/locationencoder.

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。

↑