IEEE Trans Haptics. 2021 Apr-Jun;14(2):316-321. doi: 10.1109/TOH.2021.3078889. Epub 2021 Jun 17.
We present a neural network-based compression artifact removal technique for vibrotactile signals. The proposed decoder-side quality enhancement approach is based on recurrent neural networks (RNNs) and the principle of residual learning. We use a total of 8 nonlinear RNN layers trained to first estimate the difference between the original and the compressed signal. The estimated difference signal is then added to the compressed signal, followed by further linear processing steps to construct the enhanced signal. With our approach, we are able to enhance signals at almost all compression ratios by up to $1.25\ \mathrm {dB}$. For the signals in our data set, rougly 86% are enhanced in their quality. Through an ablation study, we show that every block of our network is functioning as intended and contributes to the compression artifact removal. Additionally, we show that the chosen network parameters maximize performance.
我们提出了一种基于神经网络的压缩伪影去除技术，用于振动触觉信号。所提出的解码器端质量增强方法基于递归神经网络（RNN）和残差学习原理。我们总共使用了 8 个非线性 RNN 层进行训练，以首先估计原始信号和压缩信号之间的差异。然后，将估计的差分信号添加到压缩信号中，然后再进行进一步的线性处理步骤，以构建增强的信号。通过我们的方法，我们能够将信号增强到几乎所有的压缩比，最高可达 1.25dB。在我们的数据集的信号中，大约 86%的信号质量得到了提高。通过消融研究，我们表明我们网络的每一个块都按预期运行，并有助于去除压缩伪影。此外，我们还表明，所选择的网络参数最大限度地提高了性能。
