File(s) under permanent embargo
Deep convolutional feature transfer across mobile activity recognition domains, sensor modalities and locations
presentation
posted on 2023-06-09, 01:56 authored by Francisco Javier Ordonez Morales, Daniel RoggenDaniel RoggenDeep convolutional neural networks are powerful image and signal classifiers. One hypothesis is that kernels in the convolutional layers act as feature extractors, progressively highlighting more domain-specific features in upper layers of the network. Thus lower-level features might be suitable for transfer. We analyse this in wearable activity recognition by reusing kernels learned on a source domain on another target domain. We consider transfer between users, application domains, sensor modalities and sensor locations. We characterise the trade-offs of transferring various convolutional layers along model size, learning speed, recognition performance and training data. Through a novel kernel visualisation technique and comparative evaluations we identify what learned kernels are predominantly sensitive to, amongst sensor characteristics, motion dynamics and on-body placement. We demonstrate a ~17% decrease in training time at equal performance thanks to kernel transfer and we derive recommendations on when transfer is most suitable.
Funding
Is deep learning useful for wearable activity recognition?; G1460; GOOGLE
History
Publication status
- Published
File Version
- Accepted version
Publisher
ACMExternal DOI
Page range
92-99Presentation Type
- paper
Event name
20th International Symposiumon Wearable Computers (ISWC) 2016Event location
Heidelberg, GermanyEvent type
conferenceEvent date
12-16 September 2016Book title
Proceedings of International Symposium on Wearable ComputersDepartment affiliated with
- Engineering and Design Publications
Full text available
- No
Peer reviewed?
- Yes
Editors
Kai Kunze, Ulf BlankeLegacy Posted Date
2016-06-29First Compliant Deposit (FCD) Date
2016-06-28Usage metrics
Categories
No categories selectedKeywords
Licence
Exports
RefWorks
BibTeX
Ref. manager
Endnote
DataCite
NLM
DC