Devices and methods for configuring a wearable device for exercise tracking
An apparatus for configuring a wearable device for exercise tracking includes a data storage unit for storing sensor data generated by a sensor arrangement in the wearable device based on sensed movement of the wearable device and optical movement data including representations of an individual obtained from an image-based monitoring system; and a processing unit configured to: execute a movement tracker model on the sensor data to determine a first exercise activity and an associated first confidence indicator; execute an optical tracker model on the optical movement data to determine a second exercise activity and an associated second confidence indicator; process the movement tracker model based on the first confidence indicator and the second confidence indicator.
1 . An apparatus for configuring a wearable device for exercise tracking, the apparatus comprising:
a data storage unit, for storing sensor data including at least one of accelerometer data, gyroscope data, or inertial measurement data, generated by a sensor arrangement in the wearable device based on sensed movement of the wearable device, and optical movement data comprising representations of an individual obtained from an image-based monitoring system; and
a processing unit configured to:
execute a movement tracker model on the sensor data to determine a first exercise activity and an associated first confidence indicator;
execute an optical tracker model on the optical movement data to determine a second exercise activity and an associated second confidence indicator;
retrain the movement tracker model to identify the second exercise activity based on the sensor data according to a comparison of the first confidence indicator and the second confidence indicator.
2 . The apparatus of claim 1 , comprising:
a wireless communication unit for receiving the sensor data and movement tracker model from the wearable device.
3 . The apparatus of claim 2 , wherein the processing unit is configured to transmit processed movement tracker model data to the wearable device using the wireless communication unit.
4 . The apparatus of claim 1 , comprising:
a data input unit for obtaining said optical movement data.
5 . The apparatus of claim 1 , wherein the processing unit is configured to:
align first time stamp data of the sensor data with second time stamp data of the optical movement data to obtain a time aligned correlation between the determined first exercise activity and the determined second exercise activity.
6 . The apparatus of claim 1 , wherein
the second exercise activity is different from the first exercise activity.
7 . The apparatus of claim 1 , wherein processing unit is configured to:
compare the first confidence indicator with the second confidence indicator,
wherein the movement tracker model is retrained responsive to the second confidence indicator indicating a higher confidence than the first confidence indicator.
8 . The apparatus of claim 1 , wherein
the movement tracker model is retrained responsive to the second confidence indicator indicating a higher confidence than a prestored confidence indicator.
9 . The apparatus of claim 1 , wherein each confidence indicator comprises a probability score.
10 . A system for configuring a wearable device for exercise tracking, the exercise processing device comprising:
an image-based monitoring system configured to generate a time series of representations of an individual in the gym environment, wherein each of the representations defines a pose of the individual and comprises positions of a plurality of predefined feature points of the individual in a coordinate system; and
the apparatus according to claim 1 .
11 . The apparatus of claim 1 , wherein, when the second confidence indicator indicating a confidence higher than a prestored threshold value, the processing unit is further configured to retrain the movement tracker model to identify the second exercise activity based on the sensor data even if the first confidence indicator indicates a higher confidence than the second confidence indicator.
12 . A method for configuring a wearable device for exercise tracking, comprising:
receiving sensor data including at least one of accelerometer data, gyroscope data, or inertial measurement data, generated by a sensor arrangement in the wearable device based on sensed movement of the wearable device;
determining, by running a movement tracker model on the sensor data, a first exercise activity and an associated first confidence indicator;
obtaining optical movement data from an image-based monitoring system detecting representations of an individual, concurrently with the sensor arrangement obtaining the sensor data;
determining, by running an optical tracker model on the optical movement data, a second exercise activity and an associated second confidence indicator; and
retraining the movement tracker model to identify the second exercise activity based on the sensor data according to a comparison of the first confidence indicator and the second confidence indicator.
13 . The method of claim 12 , comprising:
receiving the movement tracker model in an apparatus by upload from the wearable device, wherein the movement tracker and the optical tracker model are run in the apparatus.
14 . The method of claim 13 , comprising:
downloading the processed movement tracker model to the wearable device.
15 . The method of claim 12 , comprising:
aligning first time stamp data of the sensor data with second time stamp data of the optical movement data to obtain a time aligned correlation between the determined first exercise activity and the determined second exercise activity.
16 . The method of claim 12 , wherein
the second exercise activity is different from the first exercise activity.
17 . The method of claim 12 ,
further comprising comparing the first confidence indicator with the second confidence indicator;
wherein the movement tracker model is retrained responsive to the second confidence indicator indicating a higher confidence than the first confidence indicator.
18 . The method of claim 12 , wherein processing the tracker model comprises:
wherein the movement tracker model is retrained responsive to the second confidence indicator indicating a higher confidence than a prestored confidence indicator.
19 . The method of claim 12 , wherein each confidence indicator comprises a probability score.
20 . A wearable device for exercise tracking, comprising:
a sensor arrangement configured to generate sensor data, including at least one of accelerometer data, gyroscope data, or inertial measurement data, based on sensed movement of the wearable device;
a data storage unit for storing the sensor data and a movement tracker model;
a wireless transceiver unit; and
a processing unit configured to execute the movement tracker model on the sensor data to determine a first exercise activity and a first confidence indicator;
wherein the processing unit is further configured to control the wireless transceiver unit to:
transmit the movement tracker model to an apparatus for configuring the wearable device for exercise tracking;
transmit the sensor data, the first exercise activity and the first confidence indicator to the apparatus;
receive processed movement tracker model data from the apparatus obtained by retraining the movement tracker model based on a comparison of the first exercise activity and first confidence indicator together with a second exercise activity and a second confidence indicator based on optical movement data from an image-based monitoring system detecting representations of an individual concurrently with the sensor arrangement obtaining the sensor data; and
wherein the processing unit is configured to:
update the movement tracker model on the in the data storage unit based on the processed movement tracker model data, wherein the updated movement tracker model identifies the second exercise activity based on the sensor data.