Masked reconstruction based self-supervision for human activity recognition
Authors:
Open Access
- 4 September 2020
- conference paper
- Published by Association for Computing Machinery (ACM)
- p. 45-49
Abstract
The ubiquitous availability of wearable sensing devices has rendered large scale collection of movement data a straightforward endeavor. Yet, annotation of these data remains a challenge and as such, publicly available datasets for human activity recognition (HAR) are typically limited in size as well as in variability, which constrains HAR model training and effectiveness. We introduce masked reconstruction as a viable self-supervised pre-training objective for human activity recognition and explore its effectiveness in comparison to state-of-the-art unsupervised learning techniques. In scenarios with small labeled datasets, the pre-training results in improvements over end-to-end learning on two of the four benchmark datasets. This is promising because the pre-training objective can be integrated "as is" into state-of-the-art recognition pipelines to effectively facilitate improved model robustness, and thus, ultimately, leading to better recognition performance.Keywords
This publication has 17 references indexed in Scilit:
- Colorization as a Proxy Task for Visual Understanding2017
- Shuffle and Learn: Unsupervised Learning Using Temporal Order Verification2016
- Context Encoders: Feature Learning by Inpainting2016
- Deep Convolutional and LSTM Recurrent Neural Networks for Multimodal Wearable Activity RecognitionSensors, 2016
- Using unlabeled data in a sparse-coding framework for human activity recognitionPervasive and Mobile Computing, 2014
- Glove: Global Vectors for Word Representation2014
- A tutorial on human activity recognition using body-worn inertial sensorsACM Computing Surveys, 2014
- On preserving statistical characteristics of accelerometry data using their empirical cumulative distribution2013
- Preprocessing techniques for context recognition from accelerometer dataPersonal and Ubiquitous Computing, 2010
- Extracting and composing robust features with denoising autoencoders2008