UTCS Professor Kristen Grauman received the 2017 Helmholtz Prize last week at the International Conference on Computer Vision. The Helmholtz Prize, awarded every other year, recognizes ICCV papers from the past ten years that have had a significant impact on the field of computer vision research.

The award is for a paper that Grauman co-authored and presented at ICCV in 2005, “The pyramid match kernel: Discriminative classification with sets of image features.” The paper discusses a method to match two different sets of features from different images in order to perform computer vision tasks such as measuring similarity, learning object categories or inferring 3-D poses. Their matching method is not sensitive to image clutter and can better handle images of objects with different backgrounds and from different viewpoints.

News categories: