Effects of environmental colour on mood: a wearable LifeColour capture device
Proceedings of the international conference on Multimedia
Passively recognising human activities through lifelogging
Computers in Human Behavior
Hi-index | 0.00 |
The SenseCam data can be used to estimate time spent in specific episodes of sedentary behaviors, as well as some dimensions of sedentary behaviors. However, it is unknown whether SenseCam data can be aggregated to provide an objective estimate of total sedentary time accumulated during a single day. We compared SenseCam-derived day-level estimates to self-report estimates of time spent in sedentary behaviors using 39 days of concurrent SenseCam and self-report data from a sample of university employed adults (age 18--70 years). We also examined whether SenseCam data can be used to compute day-level estimates of specific dimensions of sedentary behavior (e.g., co-occurring sedentary behaviors and social context). Twenty-four percent of the days of SenseCam image data collected did not have enough image data (i.e., ≥8 hours of data) to generate day-level estimates. Further, the day-level agreement between the SenseCam and self-report estimates of time spent in sedentary behaviors varied considerably by device wear time. In terms of dimensions of sedentary behaviors measured by the SenseCam, over one-third of the total sedentary time involved a social interaction and the majority (71%) of the estimated sedentary time was spent in one behavior. Overall, SenseCam data can be used to compute day-level estimates of time spent in specific episodes of sedentary behaviors and the images provide data on critical dimensions of these behaviors; however, device wear-time significantly influences the accuracy of day-level estimates.