Towards Environment Independent Device Free Human Activity Recognition

By Wenjun Jiang, Chenglin Miao, Fenglong Ma, Shuochao Yao, Yaqing Wang, Ye Yuan, Hongfei Xue, Chen Song, Xin Ma, Dimitrios Koutsonikolas and 2 more. ACM/IEEE International Conference on Mobile Computing and Networking, 2018

543 citations recorded by the source corpus.

Driven by a wide range of real-world applications, significant efforts have recently been made to explore device-free human activity recognition techniques that utilize the information collected by various wireless infrastructures to infer human activities without the need for the monitored subject to carry a dedicated device. Existing device free human activity recognition approaches and systems, though yielding reasonably good performance in certain cases, are faced with a major challenge. The wireless signals arriving at the receiving devices usually carry substantial information that is specific to the environment where the activities are recorded and the human subject who conducts the activities. Due to this reason, an activity recognition model that is trained on a specific subject in a specific environment typically does not work well when being applied to predict another subject's activities that are recorded in a different environment. To address this challenge, in this paper, we propose EI, a deep-learning based device free activity recognition framework that can remove the environment and subject specific information contained in the activity data and extract environment/subject-independent features shared by the data collected on different subjects under different environments. We conduct extensive experiments on four different device free activity recognition testbeds: WiFi, ultrasound, 60 GHz mmWave, and visible light. The experimental results demonstrate the superior effectiveness and generalizability of the proposed EI framework.

Subfields: Biomedical Engineering, Computer Vision and Pattern Recognition, Electrical and Electronic Engineering

View Paper: https://www.semanticscholar.org/paper/111b322bfbaa9450e47f4db78ca21074636c2651

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Biomedical Engineering

1. Kun Qian
2. Gang Zhou
3. Yunhao Liu
4. Guidong Zhang
5. Chen Chen
6. Yue Zheng
7. Yi Zhang
8. Youfang Lin

In Computer Vision and Pattern Recognition

1. Lu Su
2. Kun Qian
3. Zhiwen Yu
4. Bin Guo
5. Gang Zhou
6. Yunhao Liu
7. Lina Yao
8. Guidong Zhang

In Electrical and Electronic Engineering

1. Kun Qian
2. Zheng Yang
3. Gang Zhou
4. Chenshu Wu
5. Yunhao Liu
6. Guidong Zhang
7. Yi Zhang
8. Yongsen Ma

Citations in the corpus

• Zero-Effort Cross-Domain Gesture Recognition with Wi-Fi (2019, methodology): “For example, recent works [20, 50] borrow the ideas from machine learning, such as transfer learning and adversarial learning, and apply advanced learning methodologies to improve cross-domain recognition performance.”
• Cross-Domain WiFi Sensing with Channel State Information: A Survey (2022, methodology): “However, they relieve but do not get rid of the pressure of data collection because they still need a few labeled or unlabeled samples in a new testing domain [4, 5, 13, 20, 23, 25, 33, 44, 60, 65, 69, 80, 103, 107, 115, 126, 128].”
• Deep Learning for Sensor-based Human Activity Recognition (2020, background): “uilding classification models for device-free activity recognition, very few works focus on how to get equally accurate recognition performance when sensors are configured in the wild. One example is [66], where an adversarial network is incorporated with deep feature extraction models to remove the”
• Towards 3D human pose construction using wifi (2020, background): “To achieve cross-domain generalization, one possible solution is to add an adversarial neural network to the proposed deep learning framework to extract domain independent features, as in existing work on domain independent activity recognition [20].”
• WiFi Sensing with Channel State Information (2019, methodology): “EI [39] NR: Hampel Filter; ST: FFT; SE: Thresholding L: Correlation, CNN Activity Recognition Accuracy: <75% (10 users, 6 activities, 3 rooms)”
• Widar3.0: Zero-Effort Cross-Domain Gesture Recognition With Wi-Fi (2021, methodology): “For example, recent works [14], [15] borrow the ideas from recent advances in deep learning, such as transfer learning and adversarial learning, and apply advanced learning methodologies to improve cross-domain recognition performance.”