Scientists at the University of Sussex have invented a new algorithm that enables smartwatches to detect and record your every move, without being told beforehand what to look for.
Current smartwatches can recognise a limited number of particular activities, including yoga and running, but these are programmed in advance.
This new method enables the technology to discover activities as they happen, not just simply when exercising, but also when brushing your teeth or cutting vegetables. The algorithm can even track sedentary activity, for instance whether you are lying or sitting down.
Dr Hristijan Gjoreski of the University of Sussex said: “Current activity-recognition systems usually fail because they are limited to recognising a predefined set of activities, whereas of course human activities are not limited and change with time.
“Here we present a new machine-learning approach that detects new human activities as they happen in real time, and which outperforms competing approaches.”
Traditional models ‘cluster’ together bursts of activity to estimate what a person has been doing, and for how long. For example, a series of continuous steps may be clustered into a walk. Where they falter is that they do not account for pauses or interruptions in the activity, and, so, a walk interrupted with two short stops would be clustered into three separate walks.
The new algorithm tracks ongoing activity, paying close attention to transitioning, as well as the activity itself. In the example above, it assumes that the walk will continue following the short pauses, and therefore holds the data while it waits.
Dr Daniel Roggen, head of the Sensor Research Technology Group at the University of Sussex, will speak at the British Science Festival on 6 September in the event ‘In the era of wearable technologies’.
He added: “Future smartwatches will be able to better analyse and understand our activities by automatically discovering when we engage in some new type of activity.
“This new method for activity discovery paints a far richer, more accurate, picture of daily human life.
“As well as for fitness and lifestyle trackers, this can be used in healthcare scenarios and in fields such as consumer behaviour research.”
The Latest on: Activity-recognition systems
- Android phones might start authenticating you based on how you walkon December 27, 2019 at 12:00 am
Activity Recognition is a system Google first announced back at I/O 2013. It gathers sensor data from your phone to perform high-level analysis of what you’re up to; instead of an app crunching lots ...
- Master Thesis Defense: Akbar Dehghanion December 2, 2019 at 7:41 am
There is a generalized idea about the positive impact of using overlapping sliding windows on the performance of recognition systems in Human Activity Recognition. In this thesis, we analyze the ...
- An Efficient Framework Based on Segmented Block Analysis for Human Activity Recognitionon February 21, 2019 at 7:55 pm
Video surveillance systems core component is human activity recognition. Analysis and identification of human activity emphasize on understanding human behavior in the video. Human activity ...
- IMU uses activity recognition to reduce power consumptionon January 6, 2019 at 4:00 pm
The BMI270 extends system battery life by handling multiple activity tracking ... This means that gesture and activity recognition features operate in ultra-low-power mode, with current consumption of ...
- Alexa heard what you did last summer – and she knows what that was, too: AI recognizes activities from soundon October 15, 2018 at 2:21 pm
activity recognition system capable of interpreting collected sound. In other words, a software that uses devices' always-on builtin microphones to sense what exactly's going on in the background.
- Helping computers fill in the gaps between video frameson September 12, 2018 at 5:00 pm
"The system doesn't go through all the frames ... Two common CNN modules being used for activity recognition today suffer from efficiency and accuracy drawbacks. One model is accurate but ...
- Multi-Sensor Based Causal Network System for Activity Recognitionon August 5, 2018 at 5:00 pm
The aim of this project is to investigate the role of a causal network learning framework based on heterogeneous multi-sensor data and based on this to develop a causal network system for activity ...
- How to Stop Android From Sharing Your Physical Activity with Third-Party Appson October 27, 2017 at 12:13 pm
And as it turns out, Google snuck a questionable feature into the operating system with a recent update. A new permission called “activity recognition” may be tracking your physical activity and ...
- Activity monitoring for multi-inhabitant smart homeson December 24, 2014 at 3:11 pm
Activity recognition and monitoring play important roles in smart home systems, which rely on perceiving and understanding inhabitants' behavior to generate a response. Applications that require such ...
- Google Now for Android will automatically remember where you parkedon May 1, 2014 at 7:06 am
Google Now automatically detects your parking spot through Android's Activity Recognition system, a feature Google released at Google I/O 2013. Activity Recognition uses a mashup of GPS ...
via Google News and Bing News