Advances in deep learning for human activity recognition have been relatively limited due to the lack of large labelled datasets.
Recent work on predicting patient outcomes in the Intensive Care Unit (ICU) has focused heavily on the physiological time series data, largely ignoring sparse data such as diagnoses and medications.
no code implementations • 17 Dec 2020 • Christian Schroeder de Witt, Catherine Tong, Valentina Zantedeschi, Daniele De Martini, Freddie Kalaitzis, Matthew Chantry, Duncan Watson-Parris, Piotr Bilinski
Extreme precipitation events, such as violent rainfall and hail storms, routinely ravage economies and livelihoods around the developing world.
The lack of large-scale, labeled data sets impedes progress in developing robust and generalized predictive models for on-body sensor-based human activity recognition (HAR).
Overall, our work highlights the need to move away from accelerometers and calls for further exploration of using imagers for activity recognition.
In this work we present BinaryFlex, a neural network architecture that learns weighting coefficients of predefined orthogonal binary basis instead of the conventional approach of learning directly the convolutional filters.