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 Tools annual

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 monthly

Global: Zendy Plus monthly

Presents the access of premium content as premium feature

Premium Content

Global: Zendy Plus annual

Global: Zendy Free Plan

In this paper, we present iDVO (inertia-embedded deep visual odometry), aself-supervised learning based monocular visual odometry (VO) for roadvehicles. When modelling the geometric consistency within adjacent frames, mostdeep VO methods ignore the temporal continuity of the camera pose, whichresults in a very severe jagged fluctuation in the velocity curves. With theobservation that road vehicles tend to perform smooth dynamic characteristicsin most of the time, we design the inertia loss function to describe theabnormal motion variation, which assists the model to learn the consecutivenessfrom long-term camera ego-motion. Based on the recurrent convolutional neuralnetwork (RCNN) architecture, our method implicitly models the dynamics of roadvehicles and the temporal consecutiveness by the extended Long Short-TermMemory (LSTM) block. Furthermore, we develop the dynamic hard-edge mask tohandle the non-consistency in fast camera motion by blocking the boundary partand which generates more efficiency in the whole non-consistency mask. Theproposed method is evaluated on the KITTI dataset, and the results demonstratestate-of-the-art performance with respect to other monocular deep VO and SLAMapproaches.

Learning by Inertia: Self-supervised Monocular Visual Odometry for Road Vehicles