InfoMAE: Pair-Efficient Cross-Modal Alignment for Multimodal Time-Series Sensing Signals
Authors:
- Tomoyoshi KimuraTKTomoyoshi Kimura
- Xinlin LiXLXinlin Li
- Osama HannaOHOsama Hanna
- Yatong ChenYCYatong Chen
- Yizhuo ChenYCYizhuo Chen
- Denizhan KaraDKDenizhan Kara
- Tianshi WangTWTianshi Wang
- Jinyang LiJLJinyang Li
- Xiaomin OuyangXOXiaomin Ouyang
- Shengzhong LiuSLShengzhong Liu
- Mani SrivastavaMSMani Srivastava
- Suhas DiggaviSDSuhas Diggavi
Open Access
- 22 April 2025
- conference paper
- Published by Association for Computing Machinery (ACM)
- p. 3084-3095
Abstract
No abstract availableAll Related Versions
This publication has 38 references indexed in Scilit:
- Robust Contrastive Learning against Noisy Views2022
- Masked Autoencoders Are Scalable Vision Learners2022
- SimMIM: a Simple Framework for Masked Image Modeling2022
- Are Multimodal Transformers Robust to Missing Modality?2022
- Robust Audio-Visual Instance Discrimination2021
- Reconsidering Representation Alignment for Multi-view Clustering2021
- Deep multisensor learning for missing-modality all-weather mappingISPRS Journal of Photogrammetry and Remote Sensing, 2021
- Contrastive Multiview Coding2020
- Missing Modalities Imputation via Cascaded Residual Autoencoder2017
- Density-ratio matching under the Bregman divergence: a unified framework of density-ratio estimationAnnals of the Institute of Statistical Mathematics, 2011