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

As a crucial task of autonomous driving, 3D object detection has made greatprogress in recent years. However, monocular 3D object detection remains achallenging problem due to the unsatisfactory performance in depth estimation.Most existing monocular methods typically directly regress the scene depthwhile ignoring important relationships between the depth and various geometricelements (e.g. bounding box sizes, 3D object dimensions, and object poses). Inthis paper, we propose to learn geometry-guided depth estimation withprojective modeling to advance monocular 3D object detection. Specifically, aprincipled geometry formula with projective modeling of 2D and 3D depthpredictions in the monocular 3D object detection network is devised. We furtherimplement and embed the proposed formula to enable geometry-aware deeprepresentation learning, allowing effective 2D and 3D interactions for boostingthe depth estimation. Moreover, we provide a strong baseline through addressingsubstantial misalignment between 2D annotation and projected boxes to ensurerobust learning with the proposed geometric formula. Experiments on the KITTIdataset show that our method remarkably improves the detection performance ofthe state-of-the-art monocular-based method without extra data by 2.80% on themoderate test setting. The model and code will be released athttps://github.com/YinminZhang/MonoGeo.

Learning Geometry-Guided Depth via Projective Modeling for Monocular 3D Object Detection