Search

Deep learning methods have recently attracted attention in solving various tasks through their learning and determining capability for powerful features. It has been known that the absence of a pooling layer in the deep learning architectures made the network performance deteriorate significantly. Existing pooling methods assume that each value in the small region of the local feature map contributes equally to generate the pooled feature map. However, some values in the pooling region may contribute partially for generating the map. In addition, existing methods usually focused on the spatial dimension and ignored the inter-channel relationship during the pooling process. Besides, most of the pooling methods utilize a static pooling strategy which is designed based on the experts' knowledge rather than learnt from the data. This study proposes a pooling method that is able to learn from the training data by utilizing a learnable attention mechanism, called Normalized Attention Inter-Channel Pooling (NAIP). The mechanism of the pooling process mainly focuses on the most important feature region for generating the pooled feature map. The proposed method is compared with the state-of-the-art works which use graph convolutional neural network and convolutional neural network architectures for skeleton-based human action recognition and image classification tasks. The experiment demonstrates that the NAIP approach outperforms the existing methods under the same circumstances. [ABSTRACT FROM AUTHOR]

Inspired by the recent success of deep learning (DL) approaches in computer vision domain, this Letter proposes a framework to encode the sensor data into an image representation for the activity recognition task. The signal from sensors is encoded based on the Gramian Angular Field. The encoding technique increases the dimension of the data, captures a local temporal relationship in terms of temporal correlation between time intervals on the geometric interpretation, and can be easily applied to the pre‐trained DL architecture. The proposed framework is examined with respect to six popular sensor‐based activity recognition datasets. Using the authors' framework, the results show that their approach outperforms most of the state‐of‐the‐art approaches. [ABSTRACT FROM AUTHOR]