IRUS Total

CodeSLAM - Learning a Compact, Optimisable Representation for Dense Visual SLAM

File Description SizeFormat 
main.pdfAccepted version4.92 MBAdobe PDFView/Open
Title: CodeSLAM - Learning a Compact, Optimisable Representation for Dense Visual SLAM
Authors: Bloesch, M
Czarnowski, J
Clark, R
Leutenegger, S
Davison, AJ
Item Type: Conference Paper
Abstract: The representation of geometry in real-time 3D perception systems continues to be a critical research issue. Dense maps capture complete surface shape and can be augmented with semantic labels, but their high dimensionality makes them computationally costly to store and process, and unsuitable for rigorous probabilistic inference. Sparse feature-based representations avoid these problems, but capture only partial scene information and are mainly useful for localisation only. 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. Conditioning the depth map on the image allows the code to only represent aspects of the local geometry which cannot directly be predicted from the image. We explain how to learn our code representation, and demonstrate its advantageous properties in monocular SLAM.
Issue Date: 17-Dec-2018
Date of Acceptance: 19-Feb-2018
URI: http://hdl.handle.net/10044/1/58316
DOI: 10.1109/CVPR.2018.00271
Publisher: IEEE
Start Page: 2560
End Page: 2568
Journal / Book Title: 2018 IEEE/CVF Conference on Computer Vision and Pattern Recognition
Copyright Statement: © 2018 IEEE. Personal use of this material is permitted. Permission from IEEE must be obtained for all other uses, in any current or future media, including reprinting/republishing this material for advertising or promotional purposes, creating new collective works, for resale or redistribution to servers or lists, or reuse of any copyrighted component of this work in other works.
Sponsor/Funder: Dyson Technology Limited
Funder's Grant Number: PO 4500501004
Conference Name: IEEE Computer Vision and Pattern Recognition 2018
Keywords: Science & Technology
Computer Science, Artificial Intelligence
Computer Science
Publication Status: Published
Start Date: 2018-06-18
Finish Date: 2018-06-23
Conference Place: Salt Lake City, Utah, USA
Online Publication Date: 2018-12-17
Appears in Collections:Computing
Faculty of Engineering