Monocular 3D object detection (M3OD) is a significant yet inherently challenging task in
autonomous driving due to absence of explicit depth cues in a single RGB image. In this …

Abstract We present Multi-View Attentive Contextualization (MvACon) a simple yet effective
method for improving 2D-to-3D feature lifting in query-based multi-view 3D (MV3D) object …

Monocular 3D object detection usually adopts direct or hierarchical label supervision.
Recently, the distillation supervision transfers the spatial knowledge from LiDAR-or stereo …

Automatic 3D object detection using monocular cameras presents significant challenges in
the context of autonomous driving. Precise labeling of 3D object scales requires accurate …

Monocular 3D object detection (M3OD) is a significant yet inherently challenging task in
autonomous driving due to absence of implicit depth cues in a single RGB image. In this …

The integration of data from diverse sensor modalities (eg, camera and LiDAR) constitutes a
prevalent methodology within the ambit of autonomous driving scenarios. Recent …

Network details. The ODM3D cross-modal distillation framework consists of a SECOND [21]
teacher and a CaDDN [19] student. The SECOND network voxelises point clouds L within …

Technological advances have made autonomous driving more and more feasible in
common driving scenarios. Many large companies such as Waymo, Tesla, GM, and Uber …

Monocular 3D lane detection is essential for autonomous driving, but challenging due to the
inherent lack of explicit spatial information. Multi-modal approaches rely on expensive depth …