SuDA: Support-based Domain Adaptation for Sim2Real Hinge Joint Tracking with Flexible Sensors

Fang Jiawei, Haishan Song, Chengxu Zuo, Xiaoxia Gao, Xiaowei Chen, Shihui Guo, Yipeng Qin
Proceedings of the 41st International Conference on Machine Learning, PMLR 235:22042-22061, 2024.

Abstract

Flexible sensors hold promise for human motion capture (MoCap), offering advantages such as wearability, privacy preservation, and minimal constraints on natural movement. However, existing flexible sensor-based MoCap methods rely on deep learning and necessitate large and diverse labeled datasets for training. These data typically need to be collected in MoCap studios with specialized equipment and substantial manual labor, making them difficult and expensive to obtain at scale. Thanks to the high-linearity of flexible sensors, we address this challenge by proposing a novel Sim2Real solution for hinge joint tracking based on domain adaptation, eliminating the need for labeled data yet achieving comparable accuracy to supervised learning. Our solution relies on a novel Support-based Domain Adaptation method, namely SuDA, which aligns the supports of the predictive functions rather than the instance-dependent distributions between the source and target domains. Extensive experimental results demonstrate the effectiveness of our method and its superiority overstate-of-the-art distribution-based domain adaptation methods in our task.

Cite this Paper

BibTeX
@InProceedings{pmlr-v235-jiawei24a, title = {{S}u{DA}: Support-based Domain Adaptation for {S}im2{R}eal Hinge Joint Tracking with Flexible Sensors}, author = {Jiawei, Fang and Song, Haishan and Zuo, Chengxu and Gao, Xiaoxia and Chen, Xiaowei and Guo, Shihui and Qin, Yipeng}, booktitle = {Proceedings of the 41st International Conference on Machine Learning}, pages = {22042--22061}, year = {2024}, editor = {Salakhutdinov, Ruslan and Kolter, Zico and Heller, Katherine and Weller, Adrian and Oliver, Nuria and Scarlett, Jonathan and Berkenkamp, Felix}, volume = {235}, series = {Proceedings of Machine Learning Research}, month = {21--27 Jul}, publisher = {PMLR}, pdf = {https://raw.githubusercontent.com/mlresearch/v235/main/assets/jiawei24a/jiawei24a.pdf}, url = {https://proceedings.mlr.press/v235/jiawei24a.html}, abstract = {Flexible sensors hold promise for human motion capture (MoCap), offering advantages such as wearability, privacy preservation, and minimal constraints on natural movement. However, existing flexible sensor-based MoCap methods rely on deep learning and necessitate large and diverse labeled datasets for training. These data typically need to be collected in MoCap studios with specialized equipment and substantial manual labor, making them difficult and expensive to obtain at scale. Thanks to the high-linearity of flexible sensors, we address this challenge by proposing a novel Sim2Real solution for hinge joint tracking based on domain adaptation, eliminating the need for labeled data yet achieving comparable accuracy to supervised learning. Our solution relies on a novel Support-based Domain Adaptation method, namely SuDA, which aligns the supports of the predictive functions rather than the instance-dependent distributions between the source and target domains. Extensive experimental results demonstrate the effectiveness of our method and its superiority overstate-of-the-art distribution-based domain adaptation methods in our task.} }
Endnote
%0 Conference Paper %T SuDA: Support-based Domain Adaptation for Sim2Real Hinge Joint Tracking with Flexible Sensors %A Fang Jiawei %A Haishan Song %A Chengxu Zuo %A Xiaoxia Gao %A Xiaowei Chen %A Shihui Guo %A Yipeng Qin %B Proceedings of the 41st International Conference on Machine Learning %C Proceedings of Machine Learning Research %D 2024 %E Ruslan Salakhutdinov %E Zico Kolter %E Katherine Heller %E Adrian Weller %E Nuria Oliver %E Jonathan Scarlett %E Felix Berkenkamp %F pmlr-v235-jiawei24a %I PMLR %P 22042--22061 %U https://proceedings.mlr.press/v235/jiawei24a.html %V 235 %X Flexible sensors hold promise for human motion capture (MoCap), offering advantages such as wearability, privacy preservation, and minimal constraints on natural movement. However, existing flexible sensor-based MoCap methods rely on deep learning and necessitate large and diverse labeled datasets for training. These data typically need to be collected in MoCap studios with specialized equipment and substantial manual labor, making them difficult and expensive to obtain at scale. Thanks to the high-linearity of flexible sensors, we address this challenge by proposing a novel Sim2Real solution for hinge joint tracking based on domain adaptation, eliminating the need for labeled data yet achieving comparable accuracy to supervised learning. Our solution relies on a novel Support-based Domain Adaptation method, namely SuDA, which aligns the supports of the predictive functions rather than the instance-dependent distributions between the source and target domains. Extensive experimental results demonstrate the effectiveness of our method and its superiority overstate-of-the-art distribution-based domain adaptation methods in our task.
APA
Jiawei, F., Song, H., Zuo, C., Gao, X., Chen, X., Guo, S. & Qin, Y.. (2024). SuDA: Support-based Domain Adaptation for Sim2Real Hinge Joint Tracking with Flexible Sensors. Proceedings of the 41st International Conference on Machine Learning, in Proceedings of Machine Learning Research 235:22042-22061 Available from https://proceedings.mlr.press/v235/jiawei24a.html.

Related Material