MUSDA proposes hierarchical domain classifiers for multi-modality feature alignment and a prototype graph strategy for multi-source prediction fusion in unsupervised domain adaptation for 3D object detection.
Are we ready for autonomous driving? the kitti vision benchmark suite
6 Pith papers cite this work. Polarity classification is still indexing.
6
Pith papers citing it
verdicts
UNVERDICTED 6representative citing papers
Simulated coordinated IR and LiDAR spoofing achieves 85.5% success deceiving MSF perception on 400 KITTI scenes by creating consistent false 3D objects.
ParkingScenes is a new multimodal dataset of 704 structured reverse and parallel parking episodes generated in CARLA with Hybrid A* and MPC trajectories, showing better model performance than unstructured simulation data.
SPL unifies unsupervised and sparsely-supervised 3D object detection via semantic pseudo-labeling that produces bounding boxes and point labels, followed by memory-based prototype learning that mines features from both labeled and unlabeled data.
GREATEN fuses surface normals with image features via gated contextual-geometric fusion and efficient sparse attentions to cut stereo matching errors by up to 30% on real datasets when trained solely on synthetic data.
The paper introduces a safety framework for datasets in autonomous driving that uses the AI Data Flywheel and lifecycle processes to identify hazards and ensure compliance with ISO/PAS 8800.
citing papers explorer
-
MUSDA: Multi-source Multi-modality Unsupervised Domain Adaptive 3D Object Detection for Autonomous Driving
MUSDA proposes hierarchical domain classifiers for multi-modality feature alignment and a prototype graph strategy for multi-source prediction fusion in unsupervised domain adaptation for 3D object detection.
-
Cross-Modal Phantom: Coordinated Camera-LiDAR Spoofing Against Multi-Sensor Fusion in Autonomous Vehicles
Simulated coordinated IR and LiDAR spoofing achieves 85.5% success deceiving MSF perception on 400 KITTI scenes by creating consistent false 3D objects.
-
ParkingScenes: A Structured Dataset for End-to-End Autonomous Parking in Simulation Scenes
ParkingScenes is a new multimodal dataset of 704 structured reverse and parallel parking episodes generated in CARLA with Hybrid A* and MPC trajectories, showing better model performance than unstructured simulation data.
-
Unified Unsupervised and Sparsely-Supervised 3D Object Detection by Semantic Pseudo-Labeling and Prototype Learning
SPL unifies unsupervised and sparsely-supervised 3D object detection via semantic pseudo-labeling that produces bounding boxes and point labels, followed by memory-based prototype learning that mines features from both labeled and unlabeled data.
-
Geometry Reinforced Efficient Attention Tuning Equipped with Normals for Robust Stereo Matching
GREATEN fuses surface normals with image features via gated contextual-geometric fusion and efficient sparse attentions to cut stereo matching errors by up to 30% on real datasets when trained solely on synthetic data.
-
Dataset Safety in Autonomous Driving: Requirements, Risks, and Assurance
The paper introduces a safety framework for datasets in autonomous driving that uses the AI Data Flywheel and lifecycle processes to identify hazards and ensure compliance with ISO/PAS 8800.