Abstract
Investigators publish new report on Ro botics - Robotics and Automation. According to news reporting from Hangzhou, Peo ple's Republic of China, by NewsRx journalists, research stated, "How to estimat e camera pose from motion-blurred images remains a challenge for visual odometry . The blurring artifacts are inevitably caused by the exposure during camera mot ion." Financial support for this research came from Leading Goose R&D Pro gram of Zhejiang Province, China. The news correspondents obtained a quote from the research from Zhejiang Univers ity, "While current visual odometry regards them as noise, we argue that it is n ecessary to extract potential information from blur artifacts, as they contain p rior knowledge of camera motion. Base on this, we propose a blur-robust visual o dometry that improves the accuracy of camera pose estimation through exposure tr ajectory. Specifically, we first use the exposure trajectory to guide pixel matc hing between neighboring frames. The blur mask is then generated based on the ma gnitude of the exposure trajectory. The mask makes the pose module pay less atte ntion to the feature information in the severely blurred regions."
NETL
NSTL