An LSTM Recurrent Network for Step Counting
This work addresses step counting accuracy for both blind and sighted individuals, but it is incremental as it applies an existing LSTM method to a new dataset.
The paper tackled step counting for blind and sighted users using smartphone sensors, achieving a 5% overcount and undercount rate with an LSTM recurrent network trained on the WeAllWork dataset.
Smartphones with sensors such as accelerometer and gyroscope can be used as pedometers and navigators. In this paper, we propose to use an LSTM recurrent network for counting the number of steps taken by both blind and sighted users, based on an annotated smartphone sensor dataset, WeAllWork. The models were trained separately for sighted people, blind people with a long cane or a guide dog for Leave-One-Out training modality. It achieved 5% overcount and undercount rate.