[TOC]
CVPR2022 |新增数据集--V2X(首个) | DAIR-V2X: A Large-Scale Dataset for Vehicle-Infrastructure Cooperative 3D Object Detection| [论文链接] [项目主页][解读链接]
CVPR2022 |新增数据集-仿真环境里程计数据|CarlaScenes: A synthetic dataset for odometry in autonomous driving| [论文链接] [代码链接][解读链接]
CVPR2022 |新增数据集-激光雷达车道线|K-Lane: Lidar Lane Dataset and Benchmark for Urban Roads and Highways| [论文链接] [代码链接][解读链接]
CVPR2022 |新增数据集-路面湿度估计|RoadSaW: A Large-Scale Dataset for Camera-Based Road Surface and Wetness Estimation| [论文链接] [代码链接][解读链接]
CVPR2022 |新增数据集-环境动态变化|SHIFT: A Synthetic Driving Dataset for Continuous Multi-Task Domain Adaptation| [论文链接] [代码链接][解读链接]
CVPR2022 |新增数据集--密集行人多模态| STCrowd: A Multimodal Dataset for Pedestrian Perception in Crowded Scenes| [论文链接] [代码链接][解读链接]
CVPR2022 |数据集合增强--样本生成 | Local and Global GANs with Semantic-Aware Upsampling for Image Generation [论文链接][代码链接][解读链接]
CVPR2022 |端到端自动驾驶|On the Choice of Data for Efficient Training and Validation of End-to-End Driving Models| [论文链接] [代码链接][解读链接]
CVPR2022 | 端到端自动驾驶| Learning from All Vehicles| [论文链接] [代码链接][解读链接]
CVPR2022 |BEV感知地图+障碍物|Cross-view Transformers for real-time Map-view Semantic Segmentation| [论文链接] [代码链接][解读链接]
CVPR2022 |BEV障碍物投影| “The Pedestrian next to the Lamppost” Adaptive Object Graphs for Better Instantaneous Mapping | [论文链接] [代码链接][解读链接]
CVPR2022 |BEV感知车道线+障碍物|Scene Representation in Bird’s-Eye View from Surrounding Cameras with Transformers| [论文链接] [代码链接][解读链接]
CVPR2022 |BEV尺度下-基于web数据学习驾驶策略|SelfD: Self-Learning Large-Scale Driving Policies From the Web| [论文链接] [代码链接][解读链接]
CVPR2022 |异常检测综述|Anomaly Detection in Autonomous Driving: A Survey| [论文链接] [代码链接][解读链接]
CVPR2022 |异常场景语义分割|Towards Robust Semantic Segmentation of Accident Scenes via Multi-Source Mixed Sampling and Meta-Learning| [论文链接] [代码链接][解读链接]
CVPR2022 |多种天气增量学习|An Efficient Domain-Incremental Learning Approach to Drive in All Weather Conditions| [论文链接] [代码链接][解读链接]
CVPR2022 | 极端天气3D检测-点云| LiDAR Snowfall Simulation for Robust 3D Object Detection | [墙内论文] [代码链接][解读链接]
CVPR2022 |数据集增强--光照条件| SIMBAR: Single Image-Based Scene Relighting For Effective Data Augmentation For Automated Driving Vision Tasks| [论文链接] [代码链接][解读链接]
CVPR2022 |多天气数据增强| InstaFormer: Instance-Aware Image-to-Image Translation with Transformer| [论文链接] [代码链接][解读链接]
CVPR2022 |雨雾天气处理| Towards Robust Rain Removal Against Adversarial Attacks: A Comprehensive Benchmark Analysis and Beyond | [论文链接] [代码链接][解读链接]
CVPR2022 |轨迹预测--Transformer|HiVT: Hierarchical Vector Transformer for Multi-Agent Motion Prediction| [论文链接] [代码链接][解读链接]
CVPR2022 |点云数据-轨迹预测|PointMotionNet: Point-Wise Motion Learning for Large-Scale LiDAR Point Clouds Sequences| [论文链接] [代码链接][解读链接]
CVPR2022 |轨迹预测|Goal-driven Self-Attentive Recurrent Networks for Trajectory Prediction| [论文链接] [代码链接][解读链接]
CVPR2022 |轨迹预测|Importance is in your attention: agent importance prediction for autonomous driving| [论文链接] [代码链接][解读链接]
CVPR2022 |轨迹预测| End-to-End Trajectory Distribution Prediction Based on Occupancy Grid Maps| [论文链接] [代码链接][解读链接]
CVPR2022 |轨迹预测-点云| Forecasting from LiDAR via Future Object Detection | [论文链接] [代码链接][解读链接]
CVPR2022 | 轨迹预测-行人 | Adaptive Trajectory Prediction via Transferable GNN| [论文链接] [代码链接][解读链接]
CVPR2022 |轨迹预测-利用上下文信息 | Raising context awareness in motion forecasting| [论文链接] [代码链接][解读链接]
CVPR2022 |轨迹预测-场景一致 | ScePT: Scene-consistent, Policy-based Trajectory Predictions for Planning | [论文链接] [代码链接][解读链接]
CVPR2022 |视觉定位| Deep Visual Geo-localization Benchmark | [论文链接] [代码链接][解读链接]
CVPR2022 |视觉定位| Rethinking Visual Geo-localization for Large-Scale Applications| [论文链接] [代码链接][解读链接]
CVPR2022 | 稀疏地图| Long-term Visual Map Sparsification with Heterogeneous GNN | [论文链接] [代码链接][解读链接]
CVPR2022 |多模态数据融合--ViT|Multimodal Token Fusion for Vision Transformers| [论文链接] [代码链接][解读链接]
CVPR2022 | 数据融合-Lidar-Camera| DeepFusion: Lidar-Camera Deep Fusion for Multi-Modal 3D Object Detection| [论文链接] [代码链接][解读链接]
CVPR2022 | 多任务学习| Task Adaptive Parameter Sharing for Multi-Task Learning | [论文链接] [代码链接][解读链接]
CVPR2022 | 多任务网络| Controllable Dynamic Multi-Task Architectures | [论文链接] [代码链接][解读链接]
CVPR2022 |3d目标检测--点云|Focal Sparse Convolutional Networks for 3D Object Detection| [论文链接] [代码链接][解读链接]
CVPR2022 |3d目标检测--点云| OccAM’s Laser: Occlusion-based Attribution Maps for 3D Object Detectors on LiDAR Data| [论文链接] [代码链接][解读链接]
CVPR2022 |3d目标检测-图像| AutoRF: Learning 3D Object Radiance Fields from Single View Observations | [论文链接] [代码链接][解读链接]
CVPR2022 |粗标注目标检测| Towards Robust Adaptive Object Detection under Noisy Annotations| [论文链接] [代码链接][解读链接]
CVPR2022 |目标检测--毫米波雷达| Exploiting Temporal Relations on Radar Perception for Autonomous Driving| [论文链接] [代码链接][解读链接]
CVPR2022 |3d目标检测-图像| Homography Loss for Monocular 3D Object Detection| [论文链接] [代码链接][解读链接]
CVPR2022 |3d目标检测--图像+点云| CAT-Det: Contrastively Augmented Transformer for Multi-modal 3D Object Detection| [论文链接] [代码链接][解读链接]
CVPR2022 |2d目标检测| AdaMixer: A Fast-Converging Query-Based Object Detector | [论文链接] [代码链接][解读链接]
CVPR2022 | 目标检测| Multi-Granularity Alignment Domain Adaptation for Object Detection | [论文链接] [代码链接][解读链接]
CVPR2022 | 3D目标检测-图像| MonoDETR: Depth-aware Transformer for Monocular 3D Object Detection | [论文链接] [代码链接][解读链接]
CVPR2022 | 2D目标检测-视频流| Real-time Object Detection for Streaming Perception | [论文链接] [代码链接][解读链接]
CVPR2022 | 3D目标检测-图像+点云| Sparse Fuse Dense: Towards High Quality 3D Detection with Depth Completion | [论文链接] [代码链接][解读链接]
CVPR2022 | 3D目标检测-图像| MonoDTR: Monocular 3D Object Detection with Depth-Aware Transformer | [论文链接] [代码链接][解读链接]
CVPR2022 | 3D目标检测-图像| MonoJSG: Joint Semantic and Geometric Cost Volume for Monocular 3D Object Detection| [论文链接] [代码链接][解读链接]
CVPR2022 | 目标检测-点云 | Point Density-Aware Voxels for LiDAR 3D Object Detection| [论文链接] [代码链接][解读链接]
CVPR2022 | 交通标志识别| Shadows can be Dangerous: Stealthy and Effective Physical-world Adversarial Attack by Natural Phenomenon [论文链接] [代码链接][解读链接]
CVPR2022 | 点云目标检测| A Unified Query-based Paradigm for Point Cloud Understanding [墙内论文链接] [代码链接][解读链接]
CVPR2022 | 目标检测-点云| A Versatile Multi-View Framework for LiDAR-based 3D Object Detection with Guidance from Panoptic Segmentation. [论文链接] [代码链接][解读链接]
CVPR2022 | 3d目标检测-图像 | Pseudo-Stereo for Monocular 3D Object Detection in Autonomous Driving. [论文链接] [代码链接][解读链接]
CVPR2022|目标检测-点云| Embracing Single Stride 3D Object Detector with Sparse Transformer[论文链接][代码链接][解读链接]
CVPR2022 |目标检测-半监督| PseudoProp: Robust Pseudo-Label Generation for Semi-Supervised Object Detection in Autonomous Driving Systems[论文链接][代码链接][解读链接]
CVPR2022 |目标检测-多相机|MUTR3D: A Multi-camera Tracking Framework via 3D-to-2D Queries| [论文链接] [代码链接][解读链接]
CVPR2022 |目标检测/跟踪-端到端|Time3D: End-to-End Joint Monocular 3D Object Detection and Tracking for Autonomous Driving| [论文链接] [代码链接][解读链接]
CVPR2022 |行人重识别| Cloning Outfits from Real-World Images to 3D Characters for Generalizable Person Re-Identification| [论文链接] [代码链接][解读链接]
CVPR2022 | 多目标跟踪-图像| MeMOT: Multi-Object Tracking with Memory | [论文链接] [代码链接][解读链接]
CVPR2022 | 目标跟踪 | Unified Transformer Tracker for Object Tracking | [论文链接] [代码链接][解读链接]
CVPR2022|目标跟踪-点云| Beyond 3D Siamese Tracking: A Motion-Centric Paradigm for 3D Single Object Tracking in Point Clouds[论文链接][代码链接][解读链接]
CVPR2022 | 目标跟踪-LSTM| TripletTrack: 3D Object Tracking using Triplet Embeddings and LSTM[论文链接][代码链接][解读链接]
CVPR2022 |全景分割| Joint Forecasting of Panoptic Segmentations with Difference Attention| [论文链接] [代码链接][解读链接]
CVPR2022 |夜间语义分割| NightLab: A Dual-level Architecture with Hardness Detection for Segmentation at Night| [论文链接] [代码链接][解读链接]
CVPR2022 |全景分割| Panoptic, Instance and Semantic Relations: A Relational Context Encoder to Enhance Panoptic Segmentation| [论文链接] [代码链接][解读链接]
CVPR2022 |目标检测分割--自监督| Image-to-Lidar Self-Supervised Distillation for Autonomous Driving Data| [论文链接] [代码链接][解读链接]
CVPR2022 |语义分割-图像| Pin the Memory: Learning to Generalize Semantic Segmentation | [论文链接] [代码链接][解读链接]
CVPR2022 | 点云分割-弱监督| Scribble-Supervised LiDAR Semantic Segmentation| [论文链接] [代码链接][解读链接]
CVPR2022 | 实例分割 | E2EC: An End-to-End Contour-based Method for High-Quality High-Speed Instance Segmentation| [论文链接] [代码链接][解读链接]
CVPR2022| 全景分割| Bending Reality: Distortion-aware Transformers for Adapting to Panoramic Semantic Segmentation[论文链接][代码链接][解读链接]
CVPR2022 |激光雷达数据-全景分割| Proposal-free Lidar Panoptic Segmentation with Pillar-level Affinity[论文链接][代码链接][解读链接]
CVPR2022 |语义分割-图像-自监督| Performance Prediction for Semantic Segmentation by a Self-Supervised Image Reconstruction Decoder[论文链接][代码链接][解读链接]
CVPR2022 |车道线检测|ONCE-3DLanes: Building Monocular 3D Lane Detection| [论文链接] [代码链接][解读链接]
CVPR2022 | 车道线检测| Towards Driving-Oriented Metric for Lane Detection Models | [论文链接] [代码链接][解读链接]
CVPR2022 | 车道线检测 | Rethinking Efficient Lane Detection via Curve Modeling. [论文链接] [代码链接][解读链接]
CVPR2022 |顶视图车道线检测 | Reconstruct from Top View: A 3D Lane Detection Approach based on Geometry Structure Prior[论文链接] [代码链接][解读链接]
CVPR2022 |多数据集融合-车道线检测 |Multi-level Domain Adaptation for Lane Detection[论文链接] [代码链接][解读链接]
CVPR2022 |360度深度估计| HiMODE: A Hybrid Monocular Omnidirectional Depth Estimation Model| [论文链接] [代码链接][解读链接]
CVPR2022 |深度估计| P3Depth: Monocular Depth Estimation with a Piecewise Planarity Prior| [论文链接] [代码链接][解读链接]
CVPR2022 |深度估计 | NeW CRFs: Neural Window Fully-connected CRFs for Monocular Depth Estimation. [论文链接] [代码链接][解读链接]
CVPR2022 |无监督-深度估计 | H-Net: Unsupervised Attention-based Stereo Depth Estimation Leveraging Epipolar Geometry[论文链接] [代码链接][解读链接]
CVPR2022 | 车流量估计| Deformation and Correspondence Aware Unsupervised Synthetic-to-Real Scene Flow Estimation for Point Clouds | [论文链接] [代码链接][解读链接]
CVPR2022 | 目标姿态估计| EPro-PnP: Generalized End-to-End Probabilistic Perspective-n-Points for Monocular Object Pose Estimation | [论文链接] [代码链接][解读链接]
CVPR2022 | 相机姿态估计| DiffPoseNet: Direct Differentiable Camera Pose Estimation | [论文链接] [代码链接][解读链接]
CVPR2022 |神经网络可解释性| Interpretable part-whole hierarchies and conceptual-semantic relationships in neural networks. [论文链接] [代码链接][解读链接]