文件名称：CodeSLAM — Learning a Compact, Optimisable Representation for Dense Visual SLAM
文件大小：867KB
文件格式：PDF
更新时间：2021-07-24 08:27:11
SLAM 实时3D感知 Dense SLAM CVPR18
We present a new compact but dense representation of scene geometry which is conditioned on the intensity data from a single image and generated from a code consisting of a small number of parameters. We are inspired by work both on learned depth from images, and auto-encoders. Our approach is suitable for use in a keyframe-based monocular dense SLAM system: While each keyframe with a code can produce a depth map, the code can be optimised efficiently jointly with pose variables and together with the codes of overlapping keyframes to attain global consistency. Condi- tioning the depth map on the image allows the code to only represent aspects of the local geometry which cannot di- rectly be predicted from the image. We explain how to learn our code representation, and demonstrate its advantageous properties in monocular SLAM.