Nuscenes
PulseAugur coverage of Nuscenes — every cluster mentioning Nuscenes across labs, papers, and developer communities, ranked by signal.
-
Unified Map Prior Encoder enhances autonomous driving mapping and planning
Researchers have developed a Unified Map Prior Encoder (UMPE) designed to integrate diverse map data, such as HD/SD vector maps, rasterized maps, and satellite imagery, into autonomous driving systems. This encoder addr…
-
Researchers develop noise-aware training for robust 3D object detection using V2X data
Researchers have developed a new method for integrating vehicle-to-everything (V2X) communication data into 3D object detection systems for autonomous driving. This approach aims to overcome the limitations of onboard s…
-
BEV segmentation models for autonomous driving lack generalizability across datasets
A new study published on arXiv evaluates the performance of Bird's-Eye View (BEV) segmentation models used in autonomous driving. Researchers found that models trained on single datasets, like nuScenes, tend to overfit …
-
ConFusion detector achieves state-of-the-art camera-radar fusion for autonomous driving
Researchers have introduced ConFusion, a novel camera-radar fusion method for 3D object detection in autonomous driving. This approach utilizes heterogeneous query interaction, combining image, radar, and world queries …
-
New framework uses prior map data to improve camera-based 3D object detection
Researchers have developed a novel framework called DualViewMapDet for camera-only 3D object detection and tracking, particularly beneficial for autonomous driving systems that lack LiDAR sensors. This method leverages …
-
OpenVO framework enhances visual odometry with temporal awareness and geometric priors
Researchers have developed OpenVO, a new framework for open-world visual odometry that accounts for temporal dynamics and works with uncalibrated cameras. Unlike previous methods that assume fixed observation frequencie…
-
ARETE paper details new method for HD map generation using vehicle fleet data
Researchers have developed ARETE, a new method for generating High-Definition (HD) maps for autonomous driving using crowdsourced vehicle data. The approach employs a Detection Transformer (DETR) model to predict vector…
-
CLLAP framework enhances radar-camera fusion for autonomous driving with LiDAR pretraining
Researchers have developed CLLAP, a new pretraining framework that uses contrastive learning to improve radar-camera fusion for 3D object detection in autonomous driving. The method generates pseudo-radar data from abun…
-
DVGT-2 model advances autonomous driving with real-time geometry and planning
Researchers have introduced DVGT-2, a novel Vision-Geometry-Action (VGA) model designed for autonomous driving. Unlike previous vision-language-action models, DVGT-2 prioritizes dense 3D geometry for decision-making. Th…