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

Wepresentanewcompactbutdenserepresentationofscenegeometrywhichisconditionedontheintensitydatafromasingleimageandgeneratedfromacodeconsistingofasmallnumberofparameters.Weareinspiredbyworkbothonlearneddepthfromimages,andauto-encoders.Ourapproachissuitableforuseinakeyframe-basedmo