ICCV2021_SLAM_Autonomous_driving_paper_list

ICCV2021 paper list about Autonomous driving, SLAM, VO, Localization, Feature, SFM, Place recognition, Optical flow, Point cloud registration, Depth estimation, Robust estimation, etc.

Autonomous driving:

1. Structured Bird's-Eye-View Traffic Scene Understanding From Onboard Images[pdf] [supp]
2. Chen_MultiSiam_Self-Supervised_Multi-Instance_Siamese_Representation_Learning_for_Autonomous_Driving_ICCV_2021_paper [pdf] [supp] [arXiv]
3. Chitta_NEAT_Neural_Attention_Fields_for_End-to-End_Autonomous_Driving_ICCV_2021_paper[pdf] [arXiv]
4. Cui_LookOut_Diverse_Multi-Future_Prediction_and_Planning_for_Self-Driving_ICCV_2021_paper[pdf] [supp] [arXiv]
5. Das_TMCOSS_Thresholded_Multi-Criteria_Online_Subset_Selection_for_Data-Efficient_Autonomous_Driving_ICCV_2021_paper[pdf] [supp]
6. Ettinger_Large_Scale_Interactive_Motion_Forecasting_for_Autonomous_Driving_The_Waymo_ICCV_2021_paper[pdf] [supp] [arXiv]
7. Luo_Exploring_Simple_3D_Multi-Object_Tracking_for_Autonomous_Driving_ICCV_2021_paper[pdf] [supp] [arXiv]
8. Ren_Safety-Aware_Motion_Prediction_With_Unseen_Vehicles_for_Autonomous_Driving_ICCV_2021_paper[pdf] [supp] [arXiv]
9. Schon_MGNet_Monocular_Geometric_Scene_Understanding_for_Autonomous_Driving_ICCV_2021_paper[pdf] [supp]
10. Zhang_X-World_Accessibility_Vision_and_Autonomy_Meet_ICCV_2021_paper[pdf]
11. ACDC: The Adverse Conditions Dataset With Correspondences for Semantic Driving Scene Understanding[pdf] [supp] [arXiv]

SLAM:

1. Dusmanu_Cross-Descriptor_Visual_Localization_and_Mapping_ICCV_2021_paper[pdf] [supp] [arXiv]
2. Larsson_Orthographic-Perspective_Epipolar_Geometry_ICCV_2021_paper[pdf] [supp]
3. Sucar_iMAP_Implicit_Mapping_and_Positioning_in_Real-Time_ICCV_2021_paper[pdf] [supp] [arXiv]
4. Zhu_Transfusion_A_Novel_SLAM_Method_Focused_on_Transparent_Objects_ICCV_2021_paper[pdf] [supp]

VO:

1. Liu_MBA-VO_Motion_Blur_Aware_Visual_Odometry_ICCV_2021_paper[pdf] [supp]
2. The Surprising Effectiveness of Visual Odometry Techniques for Embodied PointGoal Navigation[pdf] [supp] [arXiv]

Feature & matching:

1. Berton_Viewpoint_Invariant_Dense_Matching_for_Visual_Geolocalization_ICCV_2021_paper[pdf] [supp] [arXiv]
2. Chen_Learning_To_Match_Features_With_Seeded_Graph_Matching_Network_ICCV_2021_paper[pdf] [supp] [arXiv]
3. Hong_Deep_Matching_Prior_Test-Time_Optimization_for_Dense_Correspondence_ICCV_2021_paper[pdf] [supp] [arXiv]
4. Truong_Warp_Consistency_for_Unsupervised_Learning_of_Dense_Correspondences_ICCV_2021_paper[pdf] [supp] [arXiv]
5. Wang_P2-Net_Joint_Description_and_Detection_of_Local_Features_for_Pixel_ICCV_2021_paper[pdf] [supp]
6. Zhang_ELSD_Efficient_Line_Segment_Detector_and_Descriptor_ICCV_2021_paper[pdf] [supp] [arXiv]
7. Zhang_Self-Supervised_Pretraining_of_3D_Features_on_Any_Point-Cloud_ICCV_2021_paper[pdf] [supp] [arXiv]
8. Zhao_Multi-Scale_Matching_Networks_for_Semantic_Correspondence_ICCV_2021_paper[pdf] [supp] [arXiv]

Back-end:

1. Demmel_Square_Root_Marginalization_for_Sliding-Window_Bundle_Adjustment_ICCV_2021_paper[pdf] [supp] [arXiv]
2. Li_PoGO-Net_Pose_Graph_Optimization_With_Graph_Neural_Networks_ICCV_2021_paper[pdf] [supp]
3. Lin_BARF_Bundle-Adjusting_Neural_Radiance_Fields_ICCV_2021_paper[pdf] [supp] [arXiv]
4. Tanaka_Learning_To_Bundle-Adjust_A_Graph_Network_Approach_to_Faster_Optimization_ICCV_2021_paper[pdf]

Localization:

1. Brachmann_On_the_Limits_of_Pseudo_Ground_Truth_in_Visual_Camera_ICCV_2021_paper[pdf] [supp] [arXiv]
2. Hyeon_Pose_Correction_for_Highly_Accurate_Visual_Localization_in_Large-Scale_Indoor_ICCV_2021_paper[pdf] [supp]
3. Kim_PICCOLO_Point_Cloud-Centric_Omnidirectional_Localization_ICCV_2021_paper[pdf] [supp] [arXiv]
4. Wang_Continual_Learning_for_Image-Based_Camera_Localization_ICCV_2021_paper[pdf] [supp] [arXiv]

SFM:

1. Lindenberger_Pixel-Perfect_Structure-From-Motion_With_Featuremetric_Refinement_ICCV_2021_paper[pdf] [supp] [arXiv] （最佳学生论文）
2. Iglesias_Radial_Distortion_Invariant_Factorization_for_Structure_From_Motion_ICCV_2021_paper[pdf] [supp]
3. Moran_Deep_Permutation_Equivariant_Structure_From_Motion_ICCV_2021_paper[pdf] [supp] [arXiv]
4. Shabani_Extreme_Structure_From_Motion_for_Indoor_Panoramas_Without_Visual_Overlaps_ICCV_2021_paper[pdf] [supp]
5. Zeng_PR-RRN_Pairwise-Regularized_Residual-Recursive_Networks_for_Non-Rigid_Structure-From-Motion_ICCV_2021_paper[pdf] [supp]

Optical flow:

1. Li_GyroFlow_Gyroscope-Guided_Unsupervised_Optical_Flow_Learning_ICCV_2021_paper[pdf] [arXiv]
2. Poggi_Sensor-Guided_Optical_Flow_ICCV_2021_paper[pdf] [supp] [arXiv]
3. Zhang_Separable_Flow_Learning_Motion_Cost_Volumes_for_Optical_Flow_Estimation_ICCV_2021_paper[pdf] [supp]
4. High-Resolution Optical Flow From 1D Attention and Correlation[pdf] [supp] [arXiv]

Robust estimator:

1. Aiger_Efficient_Large_Scale_Inlier_Voting_for_Geometric_Vision_Problems_ICCV_2021_paper[pdf] [arXiv]
2. Ivashechkin_VSAC_Efficient_and_Accurate_Estimator_for_H_and_F_ICCV_2021_paper[pdf] [supp] [arXiv]

Place recognition:

1. Hui_Pyramid_Point_Cloud_Transformer_for_Large-Scale_Place_Recognition_ICCV_2021_paper[pdf]
2. Peng_Attentional_Pyramid_Pooling_of_Salient_Visual_Residuals_for_Place_Recognition_ICCV_2021_paper[pdf] [supp]
3. Peng_Conformer_Local_Features_Coupling_Global_Representations_for_Visual_Recognition_ICCV_2021_paper[pdf] [supp]

Point cloud completion:

1. Huang_RFNet_Recurrent_Forward_Network_for_Dense_Point_Cloud_Completion_ICCV_2021_paper[pdf]
2. Luo_PU-EVA_An_Edge-Vector_Based_Approximation_Solution_for_Flexible-Scale_Point_Cloud_ICCV_2021_paper[pdf] [supp]
3. Xiang_SnowflakeNet_Point_Cloud_Completion_by_Snowflake_Point_Deconvolution_With_Skip-Transformer_ICCV_2021_paper[pdf] [supp] [arXiv]
4. PoinTr: Diverse Point Cloud Completion With Geometry-Aware Transformers[pdf] [supp] [arXiv]

Point cloud registration:

1. Lee_Deep_Hough_Voting_for_Robust_Global_Registration_ICCV_2021_paper[pdf] [arXiv]
2. Cao_PCAM_Product_of_Cross Attention_Matrices_for_Rigid_Registration_of_Point_ICCV_2021_paper[pdf] [supp]
3. Deng_A_Robust_Loss_for_Point_Cloud_Registration_ICCV_2021_paper[pdf] [supp] [arXiv]
4. Jiang_Sampling_Network_Guided_Cross-Entropy_Method_for_Unsupervised_Point_Cloud_Registration_ICCV_2021_paper[pdf] [arXiv]
5. Jubran_Provably_Approximated_Point_Cloud_Registration_ICCV_2021_paper[pdf] [supp]
6. Lee_DeepPRO_Deep_Partial_Point_Cloud_Registration_of_Objects_ICCV_2021_paper[pdf] [supp]
7. Liu_LSG-CPD_Coherent_Point_Drift_With_Local_Surface_Geometry_for_Point_ICCV_2021_paper[pdf] [supp]
8. Lu_HRegNet_A_Hierarchical_Network_for_Large-Scale_Outdoor_LiDAR_Point_Cloud_ICCV_2021_paper[pdf] [supp] [arXiv]
9. Min_Distinctiveness_Oriented_Positional_Equilibrium_for_Point_Cloud_Registration_ICCV_2021_paper[pdf]
10. Nie_Differentiable_Convolution_Search_for_Point_Cloud_Processing_ICCV_2021_paper[pdf] [supp] [arXiv]
11. Wu_Feature_Interactive_Representation_for_Point_Cloud_Registration_ICCV_2021_paper[pdf]
12. Xiang_Walk_in_the_Cloud_Learning_Curves_for_Point_Clouds_Shape_ICCV_2021_paper[pdf] [arXiv]
13. Xu_OMNet_Learning_Overlapping_Mask_for_Partial-to-Partial_Point_Cloud_Registration_ICCV_2021_paper[pdf] [supp] [arXiv]
14. Yang_Dynamical_Pose_Estimation_ICCV_2021_paper[pdf] [supp] [arXiv]

Calibration:

1. Lochman_BabelCalib_A_Universal_Approach_to_Calibrating_Central_Cameras_ICCV_2021_paper[pdf] [supp] [arXiv]

Dataset:

1. Jafarzadeh_CrowdDriven_A_New_Challenging_Dataset_for_Outdoor_Visual_Localization_ICCV_2021_paper[pdf] [supp] [arXiv]

Depth:

1. Bae_Estimating_and_Exploiting_the_Aleatoric_Uncertainty_in_Surface_Normal_Estimation_ICCV_2021_paper[pdf] [supp] [arXiv]
2. Chen_Revealing_the_Reciprocal_Relations_Between_Self-Supervised_Stereo_and_Monocular_Depth_ICCV_2021_paper[pdf]
3. Choi_Adaptive_Confidence_Thresholding_for_Monocular_Depth_Estimation_ICCV_2021_paper[pdf] [supp] [arXiv]
4. Huang_Fast_Light-Field_Disparity_Estimation_With_Multi-Disparity-Scale_Cost_Aggregation_ICCV_2021_paper[pdf] [supp]
5. Huynh_Boosting_Monocular_Depth_Estimation_With_Lightweight_3D_Point_Fusion_ICCV_2021_paper[pdf] [supp] [arXiv]
6. Ji_MonoIndoor_Towards_Good_Practice_of_Self-Supervised_Monocular_Depth_Estimation_for_ICCV_2021_paper[pdf] [supp] [arXiv]
7. Johari_DepthInSpace_Exploitation_and_Fusion_of_Multiple_Video_Frames_for_Structured-Light_ICCV_2021_paper[pdf] [supp]
8. Jung_Fine-Grained_Semantics-Aware_Representation_Enhancement_for_Self-Supervised_Monocular_Depth_Estimation_ICCV_2021_paper[pdf] [supp] [arXiv]
9. Kang_Learning_Efficient_Photometric_Feature_Transform_for_Multi-View_Stereo_ICCV_2021_paper[pdf] [arXiv]
10. Kim_Just_a_Few_Points_Are_All_You_Need_for_Multi-View_ICCV_2021_paper[pdf] [supp]
11. Lee_Attentive_and_Contrastive_Learning_for_Joint_Depth_and_Motion_Field_ICCV_2021_paper[pdf] [supp]
12. Lee_PatchMatch-RL_Deep_MVS_With_Pixelwise_Depth_Normal_and_Visibility_ICCV_2021_paper[pdf] [supp]
13. Li_Revisiting_Stereo_Depth_Estimation_From_a_Sequence-to-Sequence_Perspective_With_Transformers_ICCV_2021_paper[pdf] [supp] [arXiv]
14. Li_StructDepth_Leveraging_the_Structural_Regularities_for_Self-Supervised_Indoor_Depth_Estimation_ICCV_2021_paper[pdf] [supp] [arXiv]
15. Logothetis_PX-NET_Simple_and_Efficient_Pixel-Wise_Training_of_Photometric_Stereo_Networks_ICCV_2021_paper[pdf] [supp]
16. Long_Adaptive_Surface_Normal_Constraint_for_Depth_Estimation_ICCV_2021_paper[pdf] [supp] [arXiv]
17. Ma_EPP-MVSNet_Epipolar-Assembling_Based_Depth_Prediction_for_Multi-View_Stereo_ICCV_2021_paper[pdf]
18. Mostafavi_Event-Intensity_Stereo_Estimating_Depth_by_the_Best_of_Both_Worlds_ICCV_2021_paper[pdf]
19. Peng_Excavating_the_Potential_Capacity_of_Self-Supervised_Monocular_Depth_Estimation_ICCV_2021_paper[pdf] [arXiv]
20. Qu_Bayesian_Deep_Basis_Fitting_for_Depth_Completion_With_Uncertainty_ICCV_2021_paper[pdf] [supp] [arXiv]
21. Tilmon_SaccadeCam_Adaptive_Visual_Attention_for_Monocular_Depth_Sensing_ICCV_2021_paper[pdf] [supp] [arXiv]
22. Wang_Can_Scale-Consistent_Monocular_Depth_Be_Learned_in_a_Self-Supervised_Scale-Invariant_ICCV_2021_paper[pdf]
23. Wang_Domain_Adaptive_Semantic_Segmentation_With_Self-Supervised_Depth_Estimation_ICCV_2021_paper[pdf] [supp]
24. Wang_Regularizing_Nighttime_Weirdness_Efficient_Self-Supervised_Monocular_Depth_Estimation_in_the_ICCV_2021_paper[pdf] [supp] [arXiv]
25. Wei_NerfingMVS_Guided_Optimization_of_Neural_Radiance_Fields_for_Indoor_Multi-View_ICCV_2021_paper[pdf] [supp] [arXiv]
26. Wong_Unsupervised_Depth_Completion_With_Calibrated_Backprojection_Layers_ICCV_2021_paper[pdf] [supp] [arXiv]
27. Workman_Augmenting_Depth_Estimation_With_Geospatial_Context_ICCV_2021_paper [pdf] [supp] [arXiv]
28. Xu_Digging_Into_Uncertainty_in_Self-Supervised_Multi-View_Stereo_ICCV_2021_paper[pdf] [supp] [arXiv]
29. You_Towards_Interpretable_Deep_Networks_for_Monocular_Depth_Estimation_ICCV_2021_paper[pdf] [supp] [arXiv]
30. Zhao_A_Confidence-Based_Iterative_Solver_of_Depths_and_Surface_Normals_for_ICCV_2021_paper[pdf] [supp]
31. Zhou_R-MSFM_Recurrent_Multi-Scale_Feature_Modulation_for_Monocular_Depth_Estimating_ICCV_2021_paper[pdf]
32. Self-Supervised Monocular Depth Estimation for All Day Images Using Domain Separation [pdf] [arXiv]
33. DnD: Dense Depth Estimation in Crowded Dynamic Indoor Scenes[pdf] [supp] [arXiv]

Reconstruction:

1. Bednarik_Temporally-Coherent_Surface_Reconstruction_via_Metric-Consistent_Atlases_ICCV_2021_paper[pdf] [supp] [arXiv]
2. Chen_Unsupervised_Learning_of_Fine_Structure_Generation_for_3D_Point_Clouds_ICCV_2021_paper[pdf] [supp] [arXiv]
3. Choe_VolumeFusion_Deep_Depth_Fusion_for_3D_Scene_Reconstruction_ICCV_2021_paper[pdf] [supp] [arXiv]
4. Dabral_Gravity-Aware_Monocular_3D_Human-Object_Reconstruction_ICCV_2021_paper[pdf] [supp] [arXiv]
5. Hamdi_MVTN_Multi-View_Transformation_Network_for_3D_Shape_Recognition_ICCV_2021_paper[pdf] [supp] [arXiv]
6. Jiang_Focal_Frequency_Loss_for_Image_Reconstruction_and_Synthesis_ICCV_2021_paper[pdf] [supp] [arXiv]
7. Li_3D_Building_Reconstruction_From_Monocular_Remote_Sensing_Images_ICCV_2021_paper[pdf]
8. Poliarnyi_Out-of-Core_Surface_Reconstruction_via_Global_TGV_Minimization_ICCV_2021_paper[pdf] [supp] [arXiv]
9. Reed_Dynamic_CT_Reconstruction_From_Limited_Views_With_Implicit_Neural_Representations_ICCV_2021_paper[pdf] [supp] [arXiv]
10. Shi_Geometric_Granularity_Aware_Pixel-To-Mesh_ICCV_2021_paper[pdf] [supp]
11. Song_Vis2Mesh_Efficient_Mesh_Reconstruction_From_Unstructured_Point_Clouds_of_Large_ICCV_2021_paper[pdf] [supp] [arXiv]
12. Tan_PlaneTR_Structure-Guided_Transformers_for_3D_Plane_Recovery_ICCV_2021_paper[pdf] [supp] [arXiv]
13. Wang_Multi-View_3D_Reconstruction_With_Transformers_ICCV_2021_paper[pdf] [supp] [arXiv]
14. Wei_Deep_Hybrid_Self-Prior_for_Full_3D_Mesh_Generation_ICCV_2021_paper[pdf] [supp] [arXiv]
15. Xiong_In-the-Wild_Single_Camera_3D_Reconstruction_Through_Moving_Water_Surfaces_ICCV_2021_paper[pdf] [supp]
16. Zhang_Interacting_Two-Hand_3D_Pose_and_Shape_Reconstruction_From_Single_Color_ICCV_2021_paper[pdf] [supp]
17. Zhang_Learning_Signed_Distance_Field_for_Multi-View_Surface_Reconstruction_ICCV_2021_paper[pdf] [supp] [arXiv]
18. Zhu_Semantic-Embedded_Unsupervised_Spectral_Reconstruction_From_Single_RGB_Images_in_the_ICCV_2021_paper[pdf] [supp] [arXiv]
19. Planar Surface Reconstruction From Sparse Views[pdf] [arXiv]
20. Adaptive Surface Reconstruction With Multiscale Convolutional Kernels[pdf] [supp]