Decomposing a scene into geometric and semantically consistent regions | IEEE Conference Publication | IEEE Xplore