Fullie and Wiselie: A Dual-Stream Recurrent Convolutional Attention Model for Activity Recognition

TL;DR

This paper introduces RAAF, a dual-stream recurrent convolutional attention model that adaptively selects salient features from multimodal sensor data to enhance human activity recognition accuracy and efficiency.

Contribution

The paper proposes the RAAF model with a novel activity frame generation and attention mechanism, improving feature selection and recognition performance in HAR.

Findings

RAAF achieves competitive accuracy on benchmark datasets.

The model effectively reduces computational complexity.

It demonstrates robustness in real-world scenarios.

Abstract

Multimodal features play a key role in wearable sensor based Human Activity Recognition (HAR). Selecting the most salient features adaptively is a promising way to maximize the effectiveness of multimodal sensor data. In this regard, we propose a "collect fully and select wisely (Fullie and Wiselie)" principle as well as a dual-stream recurrent convolutional attention model, Recurrent Attention and Activity Frame (RAAF), to improve the recognition performance. We first collect modality features and the relations between each pair of features to generate activity frames, and then introduce an attention mechanism to select the most prominent regions from activity frames precisely. The selected frames not only maximize the utilization of valid features but also reduce the number of features to be computed effectively. We further analyze the hyper-parameters, accuracy, interpretability, and annotation dependency of the proposed model based on extensive experiments. The results show that RAAF achieves competitive performance on two benchmarked datasets and works well in real life scenarios.