Deep Learning Frameworks for Weakly-Supervised Indoor Localization

Farhad G. Zanjani, Ilia Karmanov, Hanno Ackermann, Daniel Dijkman, Simone Merlin, Ishaque Kadampot, Brian Buesker, Vamsi Vegunta, Fatih Porikli
Proceedings of the NeurIPS 2021 Competitions and Demonstrations Track, PMLR 176:349-354, 2022.

Abstract

We present two weakly-supervised deep learning frameworks for person indoor localization through Wi-Fi signal. These two frameworks, namely OT-Isomap and WiCluster, in contrast with prior works, require only room/zone level labels that is easier to acquire, compared to hard-to-acquire centimeter accuracy position labels. The OT-Isomap is a modality-agnostic model and formulates the localization problem in the context of parametric manifold learning and optimal transportation. This framework allows jointly learning a low-dimensional embedding as well as correspondences with a topological map. The WiCluster method is based on self-supervised deep clustering and metric learning models. Inspired by the deep cluster method, the Wi-Fi signals are spatially charted and represented in lower-dimensional space while a triplet margin-loss constrains an isometric representation of data on its 2D/3D intrinsic space. We demonstrate the meter-level accuracy of these two methods on both real-world Wi-Fi and camera-based indoor localization.

Cite this Paper

BibTeX
@InProceedings{pmlr-v176-zanjani22a, title = {Deep Learning Frameworks for Weakly-Supervised Indoor Localization}, author = {Zanjani, Farhad G. and Karmanov, Ilia and Ackermann, Hanno and Dijkman, Daniel and Merlin, Simone and Kadampot, Ishaque and Buesker, Brian and Vegunta, Vamsi and Porikli, Fatih}, booktitle = {Proceedings of the NeurIPS 2021 Competitions and Demonstrations Track}, pages = {349--354}, year = {2022}, editor = {Kiela, Douwe and Ciccone, Marco and Caputo, Barbara}, volume = {176}, series = {Proceedings of Machine Learning Research}, month = {06--14 Dec}, publisher = {PMLR}, pdf = {https://proceedings.mlr.press/v176/zanjani22a/zanjani22a.pdf}, url = {https://proceedings.mlr.press/v176/zanjani22a.html}, abstract = {We present two weakly-supervised deep learning frameworks for person indoor localization through Wi-Fi signal. These two frameworks, namely OT-Isomap and WiCluster, in contrast with prior works, require only room/zone level labels that is easier to acquire, compared to hard-to-acquire centimeter accuracy position labels. The OT-Isomap is a modality-agnostic model and formulates the localization problem in the context of parametric manifold learning and optimal transportation. This framework allows jointly learning a low-dimensional embedding as well as correspondences with a topological map. The WiCluster method is based on self-supervised deep clustering and metric learning models. Inspired by the deep cluster method, the Wi-Fi signals are spatially charted and represented in lower-dimensional space while a triplet margin-loss constrains an isometric representation of data on its 2D/3D intrinsic space. We demonstrate the meter-level accuracy of these two methods on both real-world Wi-Fi and camera-based indoor localization.} }
Endnote
%0 Conference Paper %T Deep Learning Frameworks for Weakly-Supervised Indoor Localization %A Farhad G. Zanjani %A Ilia Karmanov %A Hanno Ackermann %A Daniel Dijkman %A Simone Merlin %A Ishaque Kadampot %A Brian Buesker %A Vamsi Vegunta %A Fatih Porikli %B Proceedings of the NeurIPS 2021 Competitions and Demonstrations Track %C Proceedings of Machine Learning Research %D 2022 %E Douwe Kiela %E Marco Ciccone %E Barbara Caputo %F pmlr-v176-zanjani22a %I PMLR %P 349--354 %U https://proceedings.mlr.press/v176/zanjani22a.html %V 176 %X We present two weakly-supervised deep learning frameworks for person indoor localization through Wi-Fi signal. These two frameworks, namely OT-Isomap and WiCluster, in contrast with prior works, require only room/zone level labels that is easier to acquire, compared to hard-to-acquire centimeter accuracy position labels. The OT-Isomap is a modality-agnostic model and formulates the localization problem in the context of parametric manifold learning and optimal transportation. This framework allows jointly learning a low-dimensional embedding as well as correspondences with a topological map. The WiCluster method is based on self-supervised deep clustering and metric learning models. Inspired by the deep cluster method, the Wi-Fi signals are spatially charted and represented in lower-dimensional space while a triplet margin-loss constrains an isometric representation of data on its 2D/3D intrinsic space. We demonstrate the meter-level accuracy of these two methods on both real-world Wi-Fi and camera-based indoor localization.
APA
Zanjani, F.G., Karmanov, I., Ackermann, H., Dijkman, D., Merlin, S., Kadampot, I., Buesker, B., Vegunta, V. & Porikli, F.. (2022). Deep Learning Frameworks for Weakly-Supervised Indoor Localization. Proceedings of the NeurIPS 2021 Competitions and Demonstrations Track, in Proceedings of Machine Learning Research 176:349-354 Available from https://proceedings.mlr.press/v176/zanjani22a.html.

Related Material