"There is something scary about the 1st sentence of this abstract: "The wide spread usage of wearable sensors such as in smart watches has provided continuous access to valuable user generated data such as human motion that could be used to identify an individual based on his/her motion patterns such as, gait."
The conclusion is equally as scary: " Based on our experimental results, 91% subject identification accuracy was achieved using the best individual IMU and 2DTF-DCNN. We then investigated our proposed early and late sensor fusion approaches, which improved the gait identification accuracy of the system to 93.36% and 97.06%, respectively."
This is good in many ways and shows us that gait is (almost) a fingerprint, and identification systems can help in forensics, as well as determining certain gait characteristics across groups. The other side of the coin is that someone, somewhere is compiling this data and the question then becomes "Who owns this data?" and "How can I access my data?:
Good questions that we feel will probably be answered (though we may not LIKE the answer) in time. Perhaps sooner than we think...
Dehzangi O, Taherisadr M, ChangalVala R. IMU-Based Gait Recogvition Using Convolutional Neural Networks and Multi Sensor Fusion Sensors (Basel). 2017 Nov 27;17(12). pii: E2735. doi: 10.3390/s17122735.