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

The present paper reviews the classical problem of free-form curveregistration and applies it to an efficient RGBD visual odometry system calledCanny-VO, as it efficiently tracks all Canny edge features extracted from theimages. Two replacements for the distance transformation commonly used in edgeregistration are proposed: Approximate Nearest Neighbour Fields and OrientedNearest Neighbour Fields. 3D2D edge alignment benefits from these alternativeformulations in terms of both efficiency and accuracy. It removes the need forthe more computationally demanding paradigms of datato-model registration,bilinear interpolation, and sub-gradient computation. To ensure robustness ofthe system in the presence of outliers and sensor noise, the registration isformulated as a maximum a posteriori problem, and the resulting weighted leastsquares objective is solved by the iteratively re-weighted least squaresmethod. A variety of robust weight functions are investigated and the optimalchoice is made based on the statistics of the residual errors. Efficiency isfurthermore boosted by an adaptively sampled definition of the nearestneighbour fields. Extensive evaluations on public SLAM benchmark sequencesdemonstrate state-of-the-art performance and an advantage over classicalEuclidean distance fields.

Canny-VO: Visual Odometry with RGB-D Cameras based on Geometric 3D-2D Edge Alignment