English (United Kingdom)

https://curated-unify.zendy.io/wp-json/zendy-region/v1/featured_content/oa?rat=en

https://curated-unify.zendy.io/wp-json/zendy-region/v1/highlighted_journal/

Global: Zendy Plus annual

Presents the access of premium content as premium feature

Premium Content

Presents the keyphrase highlighting as premium feature

Keyphrase Highlighting

Presents the summarisation as premium feature

Summarisation

Insights

Presents the pdf analysis as premium feature

PDF Analysis

Presents the zaia usage as premium feature

ZAIA

Global: Zendy Plus monthly

Global: Zendy Tools annual

Global: Zendy Tools monthly

Global: Zendy Free Plan

High-precision lidar odomety is an essential part of autonomous driving. Inrecent years, deep learning methods have been widely used in lidar odometytasks, but most of the current methods only extract the global features of thepoint clouds. It is impossible to obtain more detailed point-level features inthis way. In addition, only the fully connected layer is used to estimate thepose. The fully connected layer has achieved obvious results in theclassification task, but the changes in pose are a continuous rather thandiscrete process, high-precision pose estimation can not be obtained only byusing the fully connected layer. Our method avoids problems mentioned above. Weuse PointPWC as our backbone network. PointPWC is originally used for sceneflow estimation. The scene flow estimation task has a strong correlation withlidar odomety. Traget point clouds can be obtained by adding the scene flow andsource point clouds. We can achieve the pose directly through ICP algorithmsolved by SVD, and the fully connected layer is no longer used. PointPWCextracts point-level features from point clouds with different sampling levels,which solves the problem of too rough feature extraction. We conductexperiments on KITTI, Ford Campus Vision and Lidar DataSe and Apollo-SouthBayDataset. Our result is comparable with the state-of-the-art unsupervised deeplearing method SelfVoxeLO.

UnPWC-SVDLO: Multi-SVD on PointPWC for Unsupervised Lidar Odometry