
Research Article
When Neural Networks Using Different Sensors Create Similar Features
2 downloads
@INPROCEEDINGS{10.1007/978-3-030-99203-3_5, author={Hugues Moreau and Andr\^{e}a Vassilev and Liming Chen}, title={When Neural Networks Using Different Sensors Create Similar Features}, proceedings={Mobile Computing, Applications, and Services. 12th EAI International Conference, MobiCASE 2021, Virtual Event, November 13--14, 2021, Proceedings}, proceedings_a={MOBICASE}, year={2022}, month={3}, keywords={Multimodal sensors Deep learning Transport Mode detection Inertial sensors Canonical Correlation Analysis}, doi={10.1007/978-3-030-99203-3_5} }
- Hugues Moreau
Andréa Vassilev
Liming Chen
Year: 2022
When Neural Networks Using Different Sensors Create Similar Features
MOBICASE
Springer
DOI: 10.1007/978-3-030-99203-3_5
Abstract
Multimodal problems are omnipresent in the real world: autonomous driving, robotic grasping, scene understanding, etc... Instead of proposing to improve an existing method or algorithm: we will use existing statistical methods to understand the features in already-existing neural networks. More precisely, we demonstrate that a fusion method relying on Canonical Correlation Analysis on features extracted from Deep Neural Networks using different sensors is equivalent to looking at the output of the networks themselves.
Copyright © 2021–2025 ICST