End-to-end autonomous driving: Challenges and frontiers
The autonomous driving community has witnessed a rapid growth in approaches that
embrace an end-to-end algorithm framework, utilizing raw sensor input to generate vehicle …
embrace an end-to-end algorithm framework, utilizing raw sensor input to generate vehicle …
Maptrv2: An end-to-end framework for online vectorized hd map construction
High-definition (HD) map provides abundant and precise static environmental information of
the driving scene, serving as a fundamental and indispensable component for planning in …
the driving scene, serving as a fundamental and indispensable component for planning in …
Sledge: Synthesizing driving environments with generative models and rule-based traffic
SLEDGE is the first generative simulator for vehicle motion planning trained on real-world
driving logs. Its core component is a learned model that is able to generate agent bounding …
driving logs. Its core component is a learned model that is able to generate agent bounding …
Fully sparse 3d occupancy prediction
Occupancy prediction plays a pivotal role in autonomous driving. Previous methods typically
construct dense 3D volumes, neglecting the inherent sparsity of the scene and suffering high …
construct dense 3D volumes, neglecting the inherent sparsity of the scene and suffering high …
Enhancing vectorized map perception with historical rasterized maps
In autonomous driving, there is growing interest in end-to-end online vectorized map
perception in bird's-eye-view (BEV) space, with an expectation that it could replace …
perception in bird's-eye-view (BEV) space, with an expectation that it could replace …
Lanemapnet: Lane network recognization and hd map construction using curve region aware temporal bird's-eye-view perception
T Zhu, J Leng, J Zhong, Z Zhang… - 2024 IEEE Intelligent …, 2024 - ieeexplore.ieee.org
The construction of local HD (High Definition) Map and Lane Network with onboard sensors
is critical for autonomous vechicles and facilitates downstream tasks. In contrast to previous …
is critical for autonomous vechicles and facilitates downstream tasks. In contrast to previous …
Topomaskv2: Enhanced instance-mask-based formulation for the road topology problem
Recently, the centerline has become a popular representation of lanes due to its advantages
in solving the road topology problem. To enhance centerline prediction, we have developed …
in solving the road topology problem. To enhance centerline prediction, we have developed …
Priormapnet: Enhancing online vectorized hd map construction with priors
Online vectorized High-Definition (HD) map construction is crucial for subsequent prediction
and planning tasks in autonomous driving. Following MapTR paradigm, recent works have …
and planning tasks in autonomous driving. Following MapTR paradigm, recent works have …
LaneCMKT: Boosting Monocular 3D Lane Detection with Cross-Modal Knowledge Transfer
Detecting 3D lane lines from monocular images is garnering increasing attention in the
Autonomous Driving (AD) area due to its cost-effective edge. However, current monocular …
Autonomous Driving (AD) area due to its cost-effective edge. However, current monocular …
Anchor3DLane++: 3D Lane Detection via Sample-Adaptive Sparse 3D Anchor Regression
In this paper, we focus on the challenging task of monocular 3D lane detection. Previous
methods typically adopt inverse perspective mapping (IPM) to transform the Front-Viewed …
methods typically adopt inverse perspective mapping (IPM) to transform the Front-Viewed …