｜ program
Program
Best paper award was presented to
ActivityGAN: Generative Adversarial Networks for Data Augmentation in
Sensor-Based Human Activity Recognition
Xi'ang Li (Huazhong University of Science and Technology, Wuhan, China)
Jinqi Luo (Nanyang Technological University, Singapore, Singapore)
Rabih Younes (Duke University, Durham, North Carolina, United States)
Proceedings
The accepted papers in HASCA workshop have been published on ACM DL.
Time is UTC(Coordinated Universal Time) on September 12, 2020.
Presentation time:
HASCA oral presentation, 12 min (10-min talk + 2-min Q&A)
SHL oral presentation, 12 min (10-min talk + 2-min Q&A)
SHL video, 1 min
Nurse oral presentation, 9 min
Nurse video, 1 min
| 1000-1130 | -Opening remarks -Nurse Challenge summary [15 min] -Nurse Challenge winner presentation [9 min] Nurse Care activity Recognition Based on Machine Learning Techniques Using Accelerometer Data. -Nurse Challenge ceremony [5 min] -SHL Challenge Introduction [2 min] -SHL Challenge Summary [15 min] Summary of the sussex-huawei locomotion-transportation recognition challenge 2020 -SHL Challenge videos broadcast [12 min] Combining LSTM and CNN for mode of transportation classification from smartphone sensors. Activity recognition for locomotion and transportation dataset using deep learning. Where are you? Human activity recognition with smartphone sensor data. Human activity recognition using multi-input CNN model with FFT spectrograms. Smartphone location identification and transport mode recognition using an ensemble of generative adversarial networks. A multi-view architecture for the SHL challenge. UPIC: user and position independent classical approach for locomotion and transportation modes recognition. Tackling the SHL recognition challenge with phone position detection and nearest neighbour smoothing. Ensemble learning for human activity recognition. Ensemble approach for sensor-based human activity recognition. Hierarchical Classification Using ML/DL for Sussex-Huawei Locomotion-Transportation (SHL) Recognition Challenge. -SHL team 1 presentation [12 min] IndRNN based long-term temporal recognition in the spatial and frequency domain. -SHL team 2 presentation [12 min] Tackling the SHL Challenge 2020 with person-specific classifiers and semi-supervised learning. -SHL team 3 presentation [12 min] DenseNetX and GRU for the Sussex-Huawei locomotion-transportation recognition challenge. -SHL ceremony (5 min) |
| 1130-1200 | Break [30 min] |
| 1200-1330 | Session Chair: Mathias Ciliberto (University of Sussex) -[HASCA] Using iOS for Inconspicuous Data Collection: A Real-World Assessment Yuuki Nishiyama, Denzil Ferreira, Wataru Sasaki, Tadashi Okoshi, Jin Nakazawa, Anind K Dey, Kaoru Sezaki -[HASCA] ActivityGANs: Generative Adversarial Networks for Data Augmentation in Sensor-Based Human Activity Recognition Xi'ang Li, Jinqi Luo, Rabih Younes -[HASCA] Improving Activity Data Collection with On-DevicePersonalization Using Fine-tuning Nattaya Mairittha, Tittaya Mairittha, Sozo Inoue -[HASCA] Social Distancing Warning System at Public Transportation by Analyzing Wi-Fi Signal from Mobile Devices Thongtat Oransirikul, Hideyuki Takada -[HASCA] Perception of Interaction between Hand and Object Yuki Toyosaka, Tsuyoshi Okita -[HASCA] MCoMat: A New Performance Metric for Imbalanced Multi-layer Activity Recognition Dataset Sayeda Shamma Alia, Paula Lago, Sozo Inoue -Nurse Challenge videos broadcast [7 min] Feature Based Random Forest Nurse Care Activity Recognition Using Accelerometer Data. A Pragmatic Signal Processing Approach for Nurse Care Activity Recognition Using Classical Machine Learning. Complex Nurse Care Activity Recognition Using Statistical Features. Nurse Care Activity Recognition Based on Convolution Neural Network for Accelerometer Data. A Window-Based Sequence-to-One Approach with Dynamic Voting Nurse Care activity Recognition Challenge: A Comparative Verification of Multiple Preprocessing Approaches. Nurse Care Activity Recognition: Using Random Forest to Handle Imbalanced Class Problem. |
| 1330-1500 | Break [90 min] |
| 1500-1630 | Session Chair: Paula Lago (Kyusyu Inst. of Tech.) -[HASCA] Action Recognition Using Spatially Distributed Radar Setup Through Microdoppler Signature Smriti Rani, Arijit Chowdhury, Andrew Gigie, Tapas Chakravarty, Arpan Pal -[HASCA]ARM Cortex M4-based Extensible Multimodal Wearable Platform for Sensor Research and Context Sensing from Motion & Sound Daniel Roggen -[HASCA] CausalBatch: Solving Complexity/Performance Tradeoffs for Deep Convolutional and LSTM Networks for Wearable Activity Recognition Lloyd Pellatt, Daniel Roggen -[HASCA] Mental stress classification during motor tasks in older adults using an Artificial Neural Network Apostolos Kalatzis, Laura Stanley, Ranjana Mehta, Rohith Karthikeyan -[HASCA] Identifying Label Noise in Time-Series Datasets Gentry Atkinson, Metsis Vangelis -Closing |