# CMD-HAR: Cross-Modal Disentanglement for Wearable Human Activity Recognition

arXiv: https://arxiv.org/abs/2503.21843 · HTML: https://mlsanity.com/paper/2503.21843/

## Key facts

- No citing papers in the map yet (data through 14 Oct 2026).
- Builds on 11 of its 39 references that are in the map.
- First posted on arXiv 27 Mar 2025.

## Builds on (most cited first, 11 of 11)

- [Two-Stream Convolutional Networks for Action Recognition in Videos](https://mlsanity.com/paper/1406.2199/) (2014)
- [Disentangling by Factorising](https://mlsanity.com/paper/1802.05983/) (2018)
- [Res2Net: A New Multi-scale Backbone Architecture](https://mlsanity.com/paper/1904.01169/) (2019)
- [Balanced Multimodal Learning via On-the-fly Gradient Modulation](https://mlsanity.com/paper/2203.15332/) (2022)
- [Boosting Multi-modal Model Performance with Adaptive Gradient Modulation](https://mlsanity.com/paper/2308.07686/) (2023)
- [P-CNN: Pose-based CNN Features for Action Recognition](https://mlsanity.com/paper/1506.03607/) (2015)
- [UniMiB SHAR: a new dataset for human activity recognition using acceleration data from smartphones](https://mlsanity.com/paper/1611.07688/) (2016)
- [Classifier-guided Gradient Modulation for Enhanced Multimodal Learning](https://mlsanity.com/paper/2411.01409/) (2024)
- [DanHAR: Dual Attention Network For Multimodal Human Activity Recognition Using Wearable Sensors](https://mlsanity.com/paper/2006.14435/) (2020)
- [Gated Res2Net for Multivariate Time Series Analysis](https://mlsanity.com/paper/2009.11705/) (2020)
- [Multi-Scale and Multi-Modal Contrastive Learning Network for Biomedical Time Series](https://mlsanity.com/paper/2312.03796/) (2023)

## Abstract (arXiv)

Human Activity Recognition (HAR) is a fundamental technology for numerous human - centered intelligent applications. Although deep learning methods have been utilized to accelerate feature extraction, issues such as multimodal data mixing, activity heterogeneity, and complex model deployment remain largely unresolved. The aim of this paper is to address issues such as multimodal data mixing, activity heterogeneity, and complex model deployment in sensor-based human activity recognition. We propose a spatiotemporal attention modal decomposition alignment fusion strategy to tackle the problem of the mixed distribution of sensor data. Key discriminative features of activities are captured through cross-modal spatio-temporal disentangled representation, and gradient modulation is combined to alleviate data heterogeneity. In addition, a wearable deployment simulation system is constructed. We conducted experiments on a large number of public datasets, demonstrating the effectiveness of the model.