Attributes' Importance for Zero-Shot Pose-Classification Based on Wearable Sensors

TL;DR

This paper introduces a method that considers attribute importance for each class in zero-shot pose classification using wearable sensors, and provides a new dataset HDPoseDS for evaluation.

Contribution

It proposes a novel attribute importance weighting approach for zero-shot learning and introduces the HDPoseDS dataset for pose classification with wearable sensors.

Findings

Achieved a 5.9% improvement over baseline methods on HDPoseDS.

Demonstrated the effectiveness of attribute importance weighting in zero-shot pose classification.

Provided the HDPoseDS dataset, the most sensor-dense wearable sensor dataset for pose recognition.

Abstract

This paper presents a simple yet effective method for improving the performance of zero-shot learning (ZSL). ZSL classifies instances of unseen classes, from which no training data is available, by utilizing the attributes of the classes. Conventional ZSL methods have equally dealt with all the available attributes, but this sometimes causes misclassification. This is because an attribute that is effective for classifying instances of one class is not always effective for another class. In this case, a metric of classifying the latter class can be undesirably influenced by the irrelevant attribute. This paper solves this problem by taking the importance of each attribute for each class into account when calculating the metric. In addition to the proposal of this new method, this paper also contributes by providing a dataset for pose classification based on wearable sensors, named HDPoseDS. It contains 22 classes of poses performed by 10 subjects with 31 IMU sensors across full body. To the best of our knowledge, it is the richest wearable-sensor dataset especially in terms of sensor density, and thus it is suitable for studying zero-shot pose/action recognition. The presented method was evaluated on HDPoseDS and outperformed relative improvement of 5.9% in comparison to the best baseline method.