An incredible new computer system is the first to be able to correctly recognize the content of a sketch with more accuracy than a human.
The breakthrough technology could lead to new ways to interact with computers using touchscreens, Queen Mary University of London reported. Sketch-a-Net is a "deep neural network," meaning it functions in a similar way to a human brain. The program pays close attention to the unique characteristics of sketches, and focuses particularly on the order in which the lines were drawn.
Using a touchscreen to sketch what once is searching for could make it easier to find items such as specific jewelry or furniture. The development could also help police when working to match an artist's impression of a suspect to a mugshot.
The program was able to identify the bird variants 'seagull', 'flying-bird', 'standing-bird' and 'pigeon' with 42.5 percent accuracy, while humans only achieved 24.8 percent accuracy. Overall, the program recognized 74.9 percent of the time compared to humans at a success rate of only 73.1 percent.
"It's exciting that our computer program can solve the task even better than humans can. Sketches are an interesting area to study because they have been used since pre-historic times for communication and now, with the increase in use of touchscreens, they are becoming a much more common communication tool again. This could really have a huge impact for areas such as police forensics, touchscreen use and image retrieval, and ultimately will help us get to the bottom of visual understanding," said Timothy Hospedales, co-author of the study and Lecturer in the School of Electronic Engineering and Computer Science, QMUL.
The research was accepted at the British Machine Vision Conference.
