Neural Motion Compression with Frequency-adaptive Fourier Feature Network
Neural Motion Compression with Frequency-adaptive Fourier Feature Network
复制标题
DOI：
10.2312/egs.20221033
复制
发表时间：
2022
期刊：
影响因子：
--
通讯作者：
Kenji Tojo;Yifei Chen;Nobuyuki Umetani
中科院分区：
文献类型：
--
作者：
Kenji Tojo;Yifei Chen;Nobuyuki Umetani
我们提出了一种基于神经网络的压缩方法，以减轻运动捕捉数据的存储成本。人类的运动，如运动，通常包括周期性运动。我们利用这种周期性，通过应用傅立叶特征的多层感知器网络。我们的新算法发现一组傅立叶特征频率的基础上的离散余弦变换（DCT）的运动。在训练过程中，我们逐渐将DCT的主频添加到当前的一组傅立叶特征频率中，直到满足给定的质量阈值。我们使用CMU运动数据集进行了实验，结果表明，我们的方法实现了整体高压缩比，同时保持其质量。
We present a neural-network-based compression method to alleviate the storage cost of motion capture data. Human motions such as locomotion, often consist of periodic movements. We leverage this periodicity by applying Fourier features to a multilayered perceptron network. Our novel algorithm finds a set of Fourier feature frequencies based on the discrete cosine transformation (DCT) of motion. During training, we incrementally added a dominant frequency of the DCT to a current set of Fourier feature frequencies until a given quality threshold was satisfied. We conducted an experiment using CMU motion dataset, and the results suggest that our method achieves overall high compression ratio while maintaining its quality.