Researchers have created an artificial intelligence algorithm capable of locating a subject appearing in surveillance footage based on description rather than on facial recognition.

According to the research team, deep learning moved "beyond machine learning (where patterns are set into algorithms and require supervision) by incorporating 'self-learning' — to train a convolutional neural network (CNN) to recognize soft biometrics using computer vision."

As such, the research team concentrated their attention on so-called soft biometrics — height, clothing (fabric color and type), build and gender.

"The task of person retrieval in the video is very challenging due to occlusion, light condition, camera quality, pose, and zoom. However, attributes like height, cloth color, gender can be deduced from low-quality surveillance video at a distance without cooperation from the subject," the authors wrote.

To demonstrate the effectiveness of the algorithm, the research team was presented with a request for footage of females wearing red shirts, standing at 153 cm tall. The algorithm correctly identified 28 subjects out of 41 in a data set of soft biometric attributes.

"The network is 8 layers deep and can classify images into 1000 object categories, such as keyboard, mouse, pencil, and many animals. As a result, the network has learned rich feature representations for a wide range of images," according to the authors.

The technology, according to researchers, could be used to help locate missing persons and to track suspected criminals.

The research team, composed of Hiren Galiyawala, Kenil Shah, Vandit Gajjar and Mehul S. Raval, detailed their work in the paper "Person Retrieval in Surveillance Video using Height, Color and Gender."