The assistant professor of computing sciences collaborated with researchers from the University of North Carolina at Charlotte on the work presented at the International Conference on Distributed Smart Cameras.
Assistant Professor of Computing Sciences Scott Spurlock, together with collaborators from the University of North Carolina at Charlotte Junjie Shan and Richard Souvenir, won the Best Conference Paper award for their work, “Discriminative Poses for Early Recognition in Multi-camera Networks,” presented at the International Conference on Distributed Smart Cameras (ICDSC) on Sept. 9 in Seville, Spain.
The research paper, which was selected from among 26 finalists as the overall best-at-conference, resulted from a multi-year collaboration. It is one of two papers authored by Spurlock to be accepted at the conference. The papers will be published in the main conference proceedings.
Spurlock, Shan and Souvenir developed an approach for the automatic recognition of common human actions, such as waving, walking or sitting. Their framework is designed to balance the accuracy of recognition with the speed of computation, to allow its use in interactive environments, such as smart-homes. Compared with prior approaches, the authors showed that their method was able to more reliably recognize actions after observing them for shorter periods of time.
Spurlock joined the Department of Computing Sciences in August after completing his doctoral degree at the University of North Carolina at Charlotte. Souvenir presented the work at the conference in Seville.