Compressive Neural Representations of Volumetric Scalar Fields
Compressive Neural Representations of Volumetric Scalar Fields
复制标题
DOI：
10.1111/cgf.14295
复制
发表时间：
2021-04
影响因子：
2.5
通讯作者：
Yuzhe Lu;K. Jiang;J. Levine;M. Berger
中科院分区：
文献类型：
--
作者：
Yuzhe Lu;K. Jiang;J. Levine;M. Berger
我们提出了一种使用隐式神经表示压缩体积标量场的方法。我们的方法将标量场表示为学习函数，其中神经网络将域中的点映射到输出标量值。通过将神经网络的权重设置为小于输入大小，我们实现了标量场的压缩表示，从而将框架压缩作为一种函数近似。结合仔细量化网络权重，我们表明这种方法产生了高度紧凑的表示，优于最先进的体积压缩方法。我们方法的概念简单性带来了许多好处，例如支持时变标量场，优化以保持空间梯度，以及随机访问字段评估。我们研究了网络设计选择对压缩性能的影响，强调了简单的网络架构对各种卷的有效性。
We present an approach for compressing volumetric scalar fields using implicit neural representations. Our approach represents a scalar field as a learned function, wherein a neural network maps a point in the domain to an output scalar value. By setting the number of weights of the neural network to be smaller than the input size, we achieve compressed representations of scalar fields, thus framing compression as a type of function approximation. Combined with carefully quantizing network weights, we show that this approach yields highly compact representations that outperform state‐of‐the‐art volume compression approaches. The conceptual simplicity of our approach enables a number of benefits, such as support for time‐varying scalar fields, optimizing to preserve spatial gradients, and random‐access field evaluation. We study the impact of network design choices on compression performance, highlighting how simple network architectures are effective for a broad range of volumes.