Researchers have developed a new method that teaches computers to understand what humans do in a typical day, an advance that may lead to devices predicting human activities and even suggesting alternatives.

The researchers at a University in US gathered more than 40,000 pictures taken every 30 to 60 seconds, over a period of six months by a wearable camera and predicted with 83% accuracy what activity that person was up to. They taught the computer to categorise images across 19 activity classes.


The test subject wearing the camera could review and annotate the photos at the end of each day to ensure that they were correctly categorised. The group believes they have gathered the largest annotated dataset of first-person images to demonstrate that deep-learning can help understand human behaviour and habits of a specific person.


Researchers believe that within the next decade, we will have ubiquitous devices that will help improve our personal choices throughout the day.