arXiv cs.AI· Zhoutong Ye, Chengwen Zhang, Zixuan Wang, Xufeng Jian, Zhe He, Qingyang Wan, Xutong Wang, Jiaqi Liu, Mingze Sun, Yu Mei, Chun Yu, Yuanchun Shi·· 12 小时前AI 评分31
IMUDINO：面向惯性感知的自蒸馏基础模型
IMUDINO: Self-distilled Foundation Model for Inertial Sensing
AI 导读
IMUDINO 是首个在人类活动识别、健康、动作捕捉与里程计等多类惯性感知任务上均取得有竞争力表现的基础模型。它把 DINO 自监督学习范式迁移到惯性数据，用保持语义的运动变换模拟运动模式与传感器配置变化，并采用带频率感知位置编码的纯 Transformer 编码器，原生支持可变输入长度与采样率。该模型学习到的全局与稠密特征可跨用户、传感器配置、任务乃至机器人里程计中的不同本体泛化。
正文 · 原文
Authors:Zhoutong Ye, Chengwen Zhang, Zixuan Wang, Xufeng Jian, Zhe He, Qingyang Wan, Xutong Wang, Jiaqi Liu, Mingze Sun, Yu Mei, Chun Yu, Yuanchun Shi
Abstract:We present IMUDINO, the first foundation model (FM) for inertial sensing to achieve competitive performance across diverse sensing tasks including human activity recognition (HAR), health, motion capture, and odometry. Prior inertial FMs often focus on one subfield of inertial sensing and sometimes underperform state-of-the-art supervised methods (e.g., in HAR), making their scope considerably narrower than FMs for vision and natural language. Moreover, they often require fixed input length and sampling rate, and generalize poorly to unseen sensor locations. To build a general-purpose foundation model for inertial sensing, we adapt the DINO self-supervised learning (SSL) paradigm to inertial data through a set of semantics-preserving motion transformations that simulate variations in motion pattern and sensor setup. For model architecture, we use a plain transformer encoder with frequency-aware positional encoding that natively supports varying input length and sampling rate. This simple setup proved remarkably effective for learning global and dense features generalizable across users, sensor setups, tasks, and, in the case of robot odometry, embodiments. We hope IMUDINO can serve as a backbone for inertial sensing and kickstart the development of general-purpose inertial FMs.
| Comments: | Official repo: this https URL |
| Subjects: | Machine Learning (cs.LG); Artificial Intelligence (cs.AI); Computer Vision and Pattern Recognition (cs.CV) |
| Cite as: | arXiv:2611.07437 [cs.LG] |
| (or arXiv:2611.07437v1 [cs.LG] for this version) | |
| https://doi.org/10.48550/arXiv.2611.07437 arXiv-issued DOI via DataCite (pending registration) |
Submission history
From: Zhoutong Ye [view email]
[v1]
Thu, 5 Nov 2026 04:35:16 UTC (12,177 KB)
来源：arXiv cs.AI · arxiv.org