自动驾驶

Fast and Efficient Transformer-based Method for Bird's Eye View Instance Prediction

快速且高效的基于变换器的鸟瞰图实例预测方法

Miguel Antunes-García, Luis M. Bergasa, Santiago Montiel-Marín, Rafael Barea, Fabio Sánchez-García, Ángel Llamazares

机构 * Universidad of Alcalá (UAH)（阿尔卡拉大学） 纠错

专题命中 感知 ：self-driving(abstract)；BEV(abstract)；occupancy(abstract)；分类 cs.CV

AI总结 本文提出一种基于简化范式的高效鸟瞰图实例预测架构，通过使用实例分割和流预测来提升速度和效率，减少参数数量和推理时间。

Comments The article has been presented in the 27th IEEE International Conference on Intelligent Transportation Systems (IEEE ITSC 2024) on September, 2024. Number of pages: 6, Number of figures: 4

Robust Dynamic Object Detection in Cluttered Indoor Scenes via Learned Spatiotemporal Cues

通过学习的时空线索在拥挤室内场景中实现鲁棒的动态物体检测

Juan Rached, Yixuan Jia, Kota Kondo, Jonathan P. How

机构 * Faculty of Electrical Engineering, Mathematics and Computer Science, University of Twente（代尔夫特理工大学电子工程、数学与计算机科学学院） ； Department of Electrical Engineering, Wright State University（怀特塞德大学电子工程系） ； Massachusetts Institute of Technology（麻省理工学院） 纠错

专题命中 感知 ：BEV(abstract)；LiDAR(abstract)；occupancy(abstract)；分类 cs.RO

AI总结 本文提出一种仅依赖LiDAR的框架，结合时间占用网格运动分割与学习的鸟瞰图动态先验，提升动态物体检测的鲁棒性，在拥挤环境中实现更高的召回率和F1分数。

RayD3D: Distilling Depth Knowledge Along the Ray for Robust Multi-View 3D Object Detection

RayD3D: 沿射线蒸馏深度知识以实现鲁棒的多视角3D目标检测

Rui Ding, Zhaonian Kuang, Zongwei Zhou, Meng Yang, Xinhu Zheng, Gang Hua

专题命中 感知 ：autonomous driving(abstract)；BEV(abstract)；LiDAR(abstract)；分类 cs.CV

AI总结 RayD3D通过沿射线蒸馏深度知识，提升多视角3D目标检测的鲁棒性，无需增加计算成本。

Fusion4CA: Boosting 3D Object Detection via Comprehensive Image Exploitation

Fusion4CA: 通过全面利用图像信息提升3D目标检测

Kang Luo, Xin Chen, Yangyi Xiao, Hesheng Wang

机构 * IRMV Lab, the Department of Automation, Shanghai Jiao Tong University（IRMV实验室，自动化系，上海交通大学） 纠错

专题命中 感知 ：autonomous driving(abstract)；BEV(abstract)；LiDAR(abstract)；分类 cs.CV

AI总结 Fusion4CA通过全面利用图像信息提升3D目标检测性能，采用对比对齐模块和相机辅助分支，结合认知适配器和坐标注意力模块，在nuScenes数据集上实现69.7%的mAP。

Learning to Drive is a Free Gift: Large-Scale Label-Free Autonomy Pretraining from Unposed In-The-Wild Videos

学习驾驶是免费礼物：从未经处理的现实视频中进行大规模无标签自主性预训练

Matthew Strong, Wei-Jer Chang, Quentin Herau, Jiezhi Yang, Yihan Hu, Chensheng Peng, Wei Zhan

机构 * Applied Intuition ； Stanford University（斯坦福大学） ； UC Berkeley（加州大学伯克利分校） 纠错

专题命中 感知 ：autonomous driving(abstract)；LiDAR(abstract)；driving perception(abstract)；分类 cs.CV

AI总结 本文提出一种无标签、教师引导的框架，通过未经处理的现实视频学习自动驾驶表示，无需姿态、标签或LiDAR，实现高效的自动驾驶感知和规划。

Comments Accepted at CVPR 2026

Easy-Poly: An Easy Polyhedral Framework For 3D Multi-Object Tracking

Easy-Poly: 一种易于使用的多目标跟踪三维多目标跟踪框架

Peng Zhang, Xin Li, Xin Lin, Liang He

机构 * East China Normal University（东华师范大学） ； Shanghai Artificial Intelligence Laboratory（上海人工智能实验室） 纠错

专题命中 感知 ：autonomous driving(abstract)；LiDAR(abstract)；driving perception(abstract)；分类 cs.CV

AI总结 Easy-Poly提出了一种基于过滤的三维多目标跟踪框架，通过四个创新方法提升小目标检测和跟踪精度，实现实时高性能表现。

Comments 8 pages, 4 figures, 6 tables

MambaFusion: Adaptive State-Space Fusion for Multimodal 3D Object Detection

MambaFusion：多模态3D目标检测的自适应状态空间融合

Venkatraman Narayanan, Bala Sai, Rahul Ahuja, Pratik Likhar, Varun Ravi Kumar, Senthil Yogamani

机构 * Qualcomm Technologies, Inc（高通技术公司） ； Qualcomm India Private Limited（高通印度私人有限公司） 纠错

专题命中 感知 ：autonomous driving(abstract)；BEV(abstract)；LiDAR(abstract)；分类 cs.CV

AI总结 MambaFusion通过结合选择性状态空间模型与窗口化变换器，实现高效且可靠的多模态3D目标检测，提升自动驾驶系统的感知能力。

AsyncBEV: Cross-modal Flow Alignment in Asynchronous 3D Object Detection

AsyncBEV: 无感异步3D目标检测中的跨模态流对齐

Shiming Wang, Holger Caesar, Liangliang Nan, Julian F. P. Kooij

专题命中 感知 ：autonomous driving(abstract)；BEV(abstract)；LiDAR(abstract)；分类 cs.CV

AI总结 AsyncBEV通过跨模态流对齐提升3D目标检测在传感器异步情况下的鲁棒性，尤其在动态物体识别中表现优异。

TopoBDA: Towards Bezier Deformable Attention for Road Topology Understanding

TopoBDA: 向贝塞尔可变形注意力迈进：道路拓扑理解

Muhammet Esat Kalfaoglu, Halil Ibrahim Ozturk, Ozsel Kilinc, Alptekin Temizel

机构 * Graduate School of Informatics, Middle East Technical University（信息学院，中东部技术大学） ； Togg/Trutek AI Team（Togg/Trutek人工智能团队） 纠错

专题命中 感知 ：autonomous driving(abstract)；BEV(abstract)；LiDAR(abstract)；分类 cs.CV

AI总结 TopoBDA通过贝塞尔可变形注意力提升道路拓扑理解，结合多模态数据增强性能。

Comments Project page: https://artest08.github.io/TopoBDA.github.io/

Journal ref Neurocomputing, Vol. 670, 132360 (2026)

Towards 3D Object-Centric Feature Learning for Semantic Scene Completion

面向语义场景补全的3D物体感知特征学习

Weihua Wang, Yubo Cui, Xiangru Lin, Zhiheng Li, Zheng Fang

专题命中 感知 ：autonomous driving(abstract)；BEV(abstract)；occupancy(abstract)；分类 cs.CV

AI总结 本文提出Ocean框架，通过分解场景为独立物体实例，提升语义场景补全的精度和性能。

Comments Accepted to AAAI-2026

Towards Sensor Data Abstraction of Autonomous Vehicle Perception Systems

迈向自动驾驶感知系统的传感器数据抽象

Hannes Reichert, Lukas Lang, Kevin Rösch, Daniel Bogdoll, Konrad Doll, Bernhard Sick, Hans-Christian Reuss, Christoph Stiller, J. Marius Zöllner

机构 * University of Kassel, Germany（卡塞尔大学） 纠错

专题命中 感知 ：autonomous driving(abstract)；LiDAR(abstract)；driving perception(abstract)；分类 cs.RO

AI总结 本文提出通过传感器数据抽象提升自动驾驶感知系统在不同传感器设置下的迁移能力。

Comments Hannes Reichert, Lukas Lang, Kevin Rösch and Daniel Bogdoll contributed equally. Accepted for publication at ISC2 2021

CATS-V2V: A Real-World Vehicle-to-Vehicle Cooperative Perception Dataset with Complex Adverse Traffic Scenarios

Hangyu Li, Bofeng Cao, Zhaohui Liang, Wuzhen Li, Juyoung Oh, Yuxuan Chen, Shixiao Liang, Hang Zhou, Chengyuan Ma, Jiaxi Liu, Zheng Li, Peng Zhang, KeKe Long, Maolin Liu, Jackson Jiang, Chunlei Yu, Shengxiang Liu, Hongkai Yu, Xiaopeng Li

机构 * University of Wisconsin-Madison（威斯康星大学麦迪逊分校） ； wuwen-ai ； Cleveland State University（克利夫兰州立大学） 纠错

专题命中 感知 ：autonomous driving(abstract)；BEV(abstract)；LiDAR(abstract)；分类 cs.CV

BEVUDA++: Geometric-aware Unsupervised Domain Adaptation for Multi-View 3D Object Detection

Rongyu Zhang, Jiaming Liu, Xiaoqi Li, Xiaowei Chi, Dan Wang, Li Du, Yuan Du, Shanghang Zhang

机构 * National Key Laboratory for Multimedia Information Processing in School of Computer Science at Peking University（北京大学计算机科学学院多媒体信息处理国家级重点实验室） ； The Hong Kong Polytechnic University（香港理工大学） ； The Hong Kong University of Science and Technology（香港科学与技术大学） ； Nanjing University（南京大学） 纠错

专题命中 感知 ：autonomous driving(abstract)；BEV(abstract)；LiDAR(abstract)；分类 cs.CV

Comments Accepted by IEEE TCSVT

SAMFusion: Sensor-Adaptive Multimodal Fusion for 3D Object Detection in Adverse Weather

Edoardo Palladin, Roland Dietze, Praveen Narayanan, Mario Bijelic, Felix Heide

机构 * Torc Robotics（Torc机器人公司） ； University of Stuttgart（斯图加特大学） ； Princeton University（普林斯顿大学） 纠错

专题命中 感知 ：autonomous driving(abstract)；BEV(abstract)；LiDAR(abstract)；分类 cs.CV

Revisiting Radar Camera Alignment by Contrastive Learning for 3D Object Detection

Linhua Kong, Dongxia Chang, Lian Liu, Zisen Kong, Pengyuan Li, Yao Zhao

机构 * IEEE Publication Technology Department（IEEE出版技术部） 纠错

专题命中 感知 ：autonomous driving(abstract)；BEV(abstract)；driving perception(abstract)；分类 cs.CV

Dual-Domain Homogeneous Fusion with Cross-Modal Mamba and Progressive Decoder for 3D Object Detection

Xuzhong Hu, Zaipeng Duan, Pei An, Jun zhang, Jie Ma

专题命中 感知 ：autonomous driving(abstract)；BEV(abstract)；LiDAR(abstract)；分类 cs.CV

Comments 13 pages, 9 figures

Progressive Multi-Modal Fusion for Robust 3D Object Detection

Rohit Mohan, Daniele Cattaneo, Florian Drews, Abhinav Valada

专题命中 感知 ：autonomous driving(abstract)；BEV(abstract)；LiDAR(abstract)；分类 cs.CV

Journal ref 8th Annual Conference on Robot Learning, 2024

Reflective Teacher: Semi-Supervised Multimodal 3D Object Detection in Bird's-Eye-View via Uncertainty Measure

Saheli Hazra, Sudip Das, Rohit Choudhary, Arindam Das, Ganesh Sistu, Ciaran Eising, Ujjwal Bhattacharya

专题命中 感知 ：autonomous driving(abstract)；BEV(abstract)；LiDAR(abstract)；分类 cs.CV

Vision-Driven 2D Supervised Fine-Tuning Framework for Bird's Eye View Perception

Lei He, Qiaoyi Wang, Honglin Sun, Qing Xu, Bolin Gao, Shengbo Eben Li, Jianqiang Wang, Keqiang Li

专题命中 感知 ：autonomous driving(abstract)；BEV(abstract)；LiDAR(abstract)；分类 cs.CV

CV-MOS: A Cross-View Model for Motion Segmentation

Xiaoyu Tang, Zeyu Chen, Jintao Cheng, Xieyuanli Chen, Jin Wu, Bohuan Xue

专题命中 感知 ：autonomous driving(abstract)；BEV(abstract)；LiDAR(abstract)；分类 cs.CV

OE-BevSeg: An Object Informed and Environment Aware Multimodal Framework for Bird's-eye-view Vehicle Semantic Segmentation

Jian Sun, Yuqi Dai, Chi-Man Vong, Qing Xu, Shengbo Eben Li, Jianqiang Wang, Lei He, Keqiang Li

专题命中 感知 ：autonomous driving(abstract)；BEV(abstract)；LiDAR(abstract)；分类 cs.CV

HeightFormer: Explicit Height Modeling without Extra Data for Camera-only 3D Object Detection in Bird's Eye View

Yiming Wu, Ruixiang Li, Zequn Qin, Xinhai Zhao, Xi Li

专题命中 感知 ：autonomous driving(abstract)；BEV(abstract)；LiDAR(abstract)；分类 cs.CV

IFTR: An Instance-Level Fusion Transformer for Visual Collaborative Perception

Shaohong Wang, Lu Bin, Xinyu Xiao, Zhiyu Xiang, Hangguan Shan, Eryun Liu

专题命中 感知 ：autonomous driving(abstract)；BEV(abstract)；LiDAR(abstract)；分类 cs.CV

DV-3DLane: End-to-end Multi-modal 3D Lane Detection with Dual-view Representation

Yueru Luo, Shuguang Cui, Zhen Li

专题命中 感知 ：autonomous driving(abstract)；BEV(abstract)；LiDAR(abstract)；分类 cs.CV

Comments Accepted at ICLR2024

RCBEVDet: Radar-camera Fusion in Bird's Eye View for 3D Object Detection

Zhiwei Lin, Zhe Liu, Zhongyu Xia, Xinhao Wang, Yongtao Wang, Shengxiang Qi, Yang Dong, Nan Dong, Le Zhang, Ce Zhu

专题命中 感知 ：autonomous driving(abstract)；BEV(abstract)；LiDAR(abstract)；分类 cs.CV

Comments Accepted by CVPR2024

CenterRadarNet: Joint 3D Object Detection and Tracking Framework using 4D FMCW Radar

Jen-Hao Cheng, Sheng-Yao Kuan, Hugo Latapie, Gaowen Liu, Jenq-Neng Hwang

专题命中 感知 ：autonomous driving(abstract)；BEV(abstract)；LiDAR(abstract)；分类 cs.CV

R2Det: Redemption from Range-view for Accurate 3D Object Detection

Yihan Wang, Qiao Yan, Yi Wang

专题命中 感知 ：autonomous driving(abstract)；BEV(abstract)；LiDAR(abstract)；分类 cs.CV

Ada3D : Exploiting the Spatial Redundancy with Adaptive Inference for Efficient 3D Object Detection

Tianchen Zhao, Xuefei Ning, Ke Hong, Zhongyuan Qiu, Pu Lu, Yali Zhao, Linfeng Zhang, Lipu Zhou, Guohao Dai, Huazhong Yang, Yu Wang

专题命中 感知 ：autonomous driving(abstract)；BEV(abstract)；LiDAR(abstract)；分类 cs.CV

Comments Accepted at ICCV2023

MonoNext: A 3D Monocular Object Detection with ConvNext

Marcelo Eduardo Pederiva, José Mario De Martino, Alessandro Zimmer

专题命中 感知 ：autonomous driving(abstract)；LiDAR(abstract)；driving perception(abstract)；分类 cs.CV

Parametric Depth Based Feature Representation Learning for Object Detection and Segmentation in Bird's Eye View

Jiayu Yang, Enze Xie, Miaomiao Liu, Jose M. Alvarez

专题命中 感知 ：autonomous driving(abstract)；BEV(abstract)；occupancy(abstract)；分类 cs.CV