Recognizing Detailed Human Context in the Wild from Smartphones and Smartwatches

By Yonatan Vaizman, Katherine Ellis, Gert R. G. Lanckriet. IEEE pervasive computing, 2016

322 citations recorded by the source corpus.

The ability to automatically recognize a person’s behavioral context can contribute to health monitoring, aging care, and many other domains. Validating context recognition in the wild is crucial to promote practical applications that work in real-life settings. The authors collected more than 300,000 minutes of sensor data with context labels from 60 subjects. Unlike previous studies, these subjects used their own personal phone, in any way that was convenient to them, and engaged in their routine in their natural environments. Unscripted behavior and unconstrained phone usage resulted in situations that were harder to recognize. The authors demonstrate how fusion of multimodal sensors is important for resolving such cases. They present a baseline system and encourage researchers to use their public dataset to compare methods and improve context recognition in the wild.

Subfields: Computer Vision and Pattern Recognition, Transportation, Human-Computer Interaction

View Paper: https://www.semanticscholar.org/paper/a85a5307c1214b9accea49e3d27bf3daa24335dc

How researchers see this paper

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

In Computer Vision and Pattern Recognition

1. Yewon Kim
2. Yong Xiang
3. Jinwoo Shin
4. Sung-Ju Lee
5. Taewon Kim
6. Shen Zhang
7. Chenhao Xu
8. Yuanbei Deng

Citations in the corpus

• NOTE: Robust Continual Test-time Adaptation Against Temporal Correlation (2022, methodology): “We evaluate NOTE under three real-world distribution datasets: object detection in autonomous driving (KITTI [9]), human activity recognition (HARTH [25]), and user behavioral context recognition (ExtraSensory [38]).”
• Asynchronous Federated Learning on Heterogeneous Devices: A Survey (2021, methodology): “[60], ExtraSensory [93] [103] A duel-weighted gradient update scheme.”
• Deep Learning in Human Activity Recognition with Wearable Sensors: A Review on Advances (2021, background): “12 100 Hz 20 ExtraSensory [279] Locomotion & activities 9D IMU & GPS 10 25–40 Hz 13 HASC [143] Locomotion Smartphone 9D IMU 6 100 Hz 11 Actitracker [299] Locomotion 9D IMU & GPS 5 N/A 6 FIC [156] Feeding gestures 3D Acc.”