2013 Volume 65 Issue 2 Pages 169-174
Recently, with the spread of on-vehicle video, it became common to share the on-vehiclevideo on website. These videos can be used for frequently and widely updating of virtual city spacesuch as Google Street View, if the location of the videos are available. To estimate the location, itis required to match the uploaded on-vehicle video to the video with location taken by a probe car.In this paper, we propose the e.cient matching method using Temporal Height Image (THI),A.neSIFT and Bag of Feature. THI retains information of relative building height from temporal imagesequence, and extract robust features by Affine SIFT. We realize e.cient matching by expressingtheir features using Bag of features. We conducted experiments to show the e.ciency of the proposedmethod by real image sequences of the city.[This abstract is not included in the PDF]