A significant gap exists in our knowledge of how domain-specific feature extraction compares to unsupervised feature learning in the latent space of a deep neural network for a range of temporal applications including human activity recognition (HAR). This paper aims to address this gap specifically for fall detection and motion recognition using acceleration data. To ensure reproducibility, we use a publicly available dataset, UniMiB-SHAR, with a well-established history in the HAR literature. We methodically analyze the performance of 64 different combinations of (i) learning representations (in the form of raw temporal data or extracted features), (ii) traditional and modern classifiers with different topologies on (iii) both binary (fall detection) and multi-class (daily activities of living) datasets. We report and discuss our findings and conclude that while feature engineering may still be competitive for HAR, trainable front-ends of modern deep learning algorithms can benefit from raw temporal data especially in large quantities. In fact, this paper claims state-of-the-art where we significantly outperform the most recent literature on this dataset in both activity recognition (88.41% vs. 98.02%) and fall detection (98.71% vs. 99.82%) using raw temporal input.

我们在特定领域特征提取与深层神经网络的潜在空间中的无监督特征学习相比（包括人类活动识别（HAR））的一系列时空应用方面的知识之间存在着巨大差距。本文旨在解决这一差距，专门用于使用加速度数据进行跌倒检测和运动识别。为了确保可重复性，我们使用了公开的数据集UniMiB-SHAR，该数据集在HAR文献中已有悠久的历史。我们系统地分析（i）学习表示（以原始时态数据或提取的特征的形式），（ii）具有不同拓扑的传统和现代分类器（iii）二进制（跌落检测）和多分类的64种不同组合的性能类（日常活动）数据集。我们报告并讨论了我们的发现，并得出结论，尽管特征工程对于HAR仍可能具有竞争力，但现代深度学习算法的可训练前端可以受益于原始时态数据，尤其是大量数据。实际上，本文采用了最新技术，在使用原始时间数据的活动识别（88.41％vs. 98.02％）和跌倒检测（98.71％vs. 99.82％）方面，我们的数据集均明显优于最新数据集输入。