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

The performance of object detection methods based on LiDAR information isheavily impacted by the availability of training data, usually limited tocertain laser devices. As a result, the use of synthetic data is becomingpopular when training neural network models, as both sensor specifications anddriving scenarios can be generated ad-hoc. However, bridging the gap betweenvirtual and real environments is still an open challenge, as current simulatorscannot completely mimic real LiDAR operation. To tackle this issue, domainadaptation strategies are usually applied, obtaining remarkable results onvehicle detection when applied to range view (RV) and bird's eye view (BEV)projections while failing for smaller road agents. In this paper, we present aBEV domain adaptation method based on CycleGAN that uses prior semanticclassification in order to preserve the information of small objects ofinterest during the domain adaptation process. The quality of the generatedBEVs has been evaluated using a state-of-the-art 3D object detection frameworkat KITTI 3D Object Detection Benchmark. The obtained results show theadvantages of the proposed method over the existing alternatives.

Cycle and Semantic Consistent Adversarial Domain Adaptation for Reducing Simulation-to-Real Domain Shift in LiDAR Bird's Eye View