English

IMU2CLIP: Multimodal Contrastive Learning for IMU Motion Sensors from Egocentric Videos and Text

Computer Vision and Pattern Recognition 2022-10-27 v1 Computation and Language Machine Learning

Abstract

We present IMU2CLIP, a novel pre-training approach to align Inertial Measurement Unit (IMU) motion sensor recordings with video and text, by projecting them into the joint representation space of Contrastive Language-Image Pre-training (CLIP). The proposed approach allows IMU2CLIP to translate human motions (as measured by IMU sensors) into their corresponding textual descriptions and videos -- while preserving the transitivity across these modalities. We explore several new IMU-based applications that IMU2CLIP enables, such as motion-based media retrieval and natural language reasoning tasks with motion data. In addition, we show that IMU2CLIP can significantly improve the downstream performance when fine-tuned for each application (e.g. activity recognition), demonstrating the universal usage of IMU2CLIP as a new pre-trained resource. Our code will be made publicly available.

Keywords

Cite

@article{arxiv.2210.14395,
  title  = {IMU2CLIP: Multimodal Contrastive Learning for IMU Motion Sensors from Egocentric Videos and Text},
  author = {Seungwhan Moon and Andrea Madotto and Zhaojiang Lin and Alireza Dirafzoon and Aparajita Saraf and Amy Bearman and Babak Damavandi},
  journal= {arXiv preprint arXiv:2210.14395},
  year   = {2022}
}