IEEE J Biomed Health Inform. 2021 Oct;25(10):3834-3843. doi: 10.1109/JBHI.2021.3092396. Epub 2021 Oct 5.
Recently, the state-of-the-art performance in various sensor based human activity recognition (HAR) tasks have been acquired by deep learning, which can extract automatically features from raw data. In order to obtain the best accuracy, many static layers have been always used to train deep neural networks, and their weight connectivity in network remains unchanged. Pursuing the best accuracy in mobile platforms with a very limited computational budget at millions of FLOPs is impractical. In this paper, we make use of shallow convolutional neural networks (CNNs) with channel-selectivity for the use of HAR. As we have known, it is for the first time to adopt channel-selectivity CNN for sensor based HAR tasks. We perform extensive experiments on 5 public benchmark HAR datasets consisting of UCI-HAR dataset, OPPORTUNITY dataset, UniMib-SHAR dataset, WISDM dataset, and PAMAP2 dataset. As a result, the channel-selectivity can achieve lower test errors than static layers. The existing performance of deep HAR can be further improved by the CNN with channel-selectivity without any extra cost.
最近，基于深度学习的最新技术在各种基于传感器的人体活动识别 (HAR) 任务中取得了优异的表现，它可以从原始数据中自动提取特征。为了获得最佳的准确性，许多静态层一直被用于训练深度神经网络，并且网络中的权重连接保持不变。在计算预算非常有限的移动平台上追求最高的准确性，以每秒浮点运算数 (FLOPs) 计达到数百万次是不切实际的。在本文中，我们利用具有通道选择性的浅层卷积神经网络 (CNN) 进行 HAR。据我们所知，这是首次将通道选择性 CNN 应用于基于传感器的 HAR 任务。我们在包含 UCI-HAR 数据集、OPPORTUNITY 数据集、UniMib-SHAR 数据集、WISDM 数据集和 PAMAP2 数据集的 5 个公共基准 HAR 数据集上进行了广泛的实验。结果表明，通道选择性可以实现比静态层更低的测试错误率。通过具有通道选择性的 CNN 可以在不增加任何额外成本的情况下进一步提高现有的 HAR 性能。
