Comparing CNN and Human Crafted Features for Human Activity Recognition
| dc.cclicence | CC-BY-NC | en |
| dc.contributor.author | Cruciani, Federico | |
| dc.contributor.author | Vafeiadis, Anastasios | |
| dc.contributor.author | Nugent, Chris | |
| dc.contributor.author | Cleland, Ian | |
| dc.contributor.author | McCullagh, Paul | |
| dc.contributor.author | Votis, Konstantinos | |
| dc.contributor.author | Giakoumis, Dimitrios | |
| dc.contributor.author | Tzovaras, Dimitrios | |
| dc.contributor.author | Chen, Liming | |
| dc.contributor.author | Hamzaoui, Raouf | |
| dc.date.acceptance | 2019-04-29 | |
| dc.date.accessioned | 2019-06-24T08:54:11Z | |
| dc.date.available | 2019-06-24T08:54:11Z | |
| dc.date.issued | 2019-08 | |
| dc.description.abstract | Deep learning techniques such as Convolutional Neural Networks (CNNs) have shown good results in activity recognition. One of the advantages of using these methods resides in their ability to generate features automatically. This ability greatly simplifies the task of feature extraction that usually requires domain specific knowledge, especially when using big data where data driven approaches can lead to anti-patterns. Despite the advantage of this approach, very little work has been undertaken on analyzing the quality of extracted features, and more specifically on how model architecture and parameters affect the ability of those features to separate activity classes in the final feature space. This work focuses on identifying the optimal parameters for recognition of simple activities applying this approach on both signals from inertial and audio sensors. The paper provides the following contributions: (i) a comparison of automatically extracted CNN features with gold standard Human Crafted Features (HCF) is given, (ii) a comprehensive analysis on how architecture and model parameters affect separation of target classes in the feature space. Results are evaluated using publicly available datasets. In particular, we achieved a 93.38% F-Score on the UCI-HAR dataset, using 1D CNNs with 3 convolutional layers and 32 kernel size, and a 90.5% F-Score on the DCASE 2017 development dataset, simplified for three classes (indoor, outdoor and vehicle), using 2D CNNs with 2 convolutional layers and a 2x2 kernel size. | en |
| dc.funder | European Union (EU) Horizon 2020 | en |
| dc.identifier.citation | F. Cruciani, A. Vafeiadis, C. Nugent, I. Cleland, P. McCullagh, K. Votis, D. Giakoumis, D. Tzovaras, L. Chen, R. Hamzaoui, Comparing CNN and human crafted features for human activity recognition. In: Proc. 16th IEEE International Conference on Ubiquitous Intelligence and Computing (UIC 2019), Leicester, Aug. 2019. | en |
| dc.identifier.uri | https://www.dora.dmu.ac.uk/handle/2086/18111 | |
| dc.language.iso | en | en |
| dc.peerreviewed | Yes | en |
| dc.projectid | ACROSSING project, Marie Skłodowska-Curie EU Framework for Research and Innovation Horizon 2020, Grant Agreement No. 676157. | en |
| dc.publisher | IEEE | en |
| dc.researchinstitute | Cyber Technology Institute (CTI) | en |
| dc.subject | Human Activity Recognition | en |
| dc.subject | Deep Learning | en |
| dc.subject | Convolutional Neural Networks | en |
| dc.title | Comparing CNN and Human Crafted Features for Human Activity Recognition | en |
| dc.type | Conference | en |