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 Tools annual

Global: Zendy Free Plan

Global: Zendy Tools monthly

Global: Zendy Plus monthly

We present a method to incorporate global orientation information from thesun into a visual odometry pipeline using only the existing image stream, wherethe sun is typically not visible. We leverage recent advances in BayesianConvolutional Neural Networks to train and implement a sun detection model thatinfers a three-dimensional sun direction vector from a single RGB image.Crucially, our method also computes a principled uncertainty associated witheach prediction, using a Monte Carlo dropout scheme. We incorporate thisuncertainty into a sliding window stereo visual odometry pipeline whereaccurate uncertainty estimates are critical for optimal data fusion. OurBayesian sun detection model achieves a median error of approximately 12degrees on the KITTI odometry benchmark training set, and yields improvementsof up to 42% in translational ARMSE and 32% in rotational ARMSE compared tostandard VO. An open source implementation of our Bayesian CNN sun estimator(Sun-BCNN) using Caffe is available at https://github.com/utiasSTARS/sun-bcnn-vo

Reducing drift in visual odometry by inferring sun direction using a Bayesian Convolutional Neural Network