Researchers at the University of Surrey developed an AI to despecific objects in a specific image and discount others. This is a “game-changer” that could lead to advancements in cancer detection. Professor Yi-Zhe Song, leading this research at the university, stated that an artist’s sketch is full of individual cues that words cannot convey concisely, reiterating the phrase a picture paints a thousand words. 

For novel AI systems, descriptive words help to generate images, but none can express the individualism of the user or the exact match the user is looking for. Yi-Zhe Song’s model is instructed by the artist via sketches to find an exact object and discount others.  

This technology is helpful in medicine for finding aggressive tumors or helping to protect wildlife conservation by detecting rare animals. In this paper, researchers share a picture of zebras, with only a sketch of a single zebra easting to direct its search.  

“The AI tool takes visual cues into account, such as pose and structure, but bases the decisions off the exact requirements given by the amateur artist”, the researchers stated in the paper.   

According to Professor Song, the ability of AI to detect objects based on individual amateur sketches introduces a significant leap in harnessing human creativity in Computer Vision. The model allows humans to interact with AI from a different perspective, no longer letting AI dictate the decisions, but asking it to behave exactly as instructed, keeping necessary human intervention. 

This research will be presented at the Computer Vision, Pattern, and Recognition Conference (CVPR) 2023, which showcases world-leading AI research on a global stage.  

This research by the University of Surrey, a research-intensive university, will be a giant leap in healthcare AI and wildlife conservation.