2021년 7월 SLAM 뉴스
논문 이름 누르면 자세한 정보가 열립니다!
이번 달 내가 관심가지는 논문들 (키노트/랜드마크 급)
CodeMapping: Real-Time Dense Mapping for Sparse SLAM using Compact Scene Representations
- [논문 링크](https://ieeexplore.ieee.org/abstract/document/9484785) - - Scene representation을 담고 있는 code (scene representation을 auto-encoder를 사용해서 압축한 정보. Optimisation을 통해 파라미터을 smooth하게 조정 가능)를 사용하여 sparse->dense reconstruction을 가능하게 함.Kimera-Multi: Robust, Distributed, Dense Metric-Semantic SLAM for Multi-Robot Systems
- [논문 링크](https://arxiv.org/pdf/2106.14386.pdf) - - 기존의 Kimera 시스템에 multi-robot (i.e. collaborative SLAM)을 적용함. - 주요 관찰점은 loop closure, map update 등이 될듯.Bootstrap Your Own Correspondences
- [논문 링크](https://arxiv.org/pdf/2106.00677.pdf) - 저스틴 존슨이요...?!DPLVO: Direct Point-Line Monocular Visual Odometry
- [논문 링크](https://ieeexplore.ieee.org/abstract/document/9484792)
LiDAR Odometry
FAST-LIO2: Fast Direct LiDAR-inertial Odometry
- [논문 링크](https://arxiv.org/pdf/2107.06829.pdf)F-LOAM : Fast LiDAR Odometry and Mapping
- [논문 링크](https://arxiv.org/pdf/2107.00822.pdf)SA-LOAM: Semantic-aided LiDAR SLAM with Loop Closure
- [논문 링크](https://arxiv.org/pdf/2106.11516.pdf)GLO-SLAM: a slam system optimally combining GPS and LiDAR odometry
- [논문 링크](https://www.emerald.com/insight/content/doi/10.1108/IR-12-2020-0272/full/html)
Visual SLAM / VINS
A Comparison of Modern General-Purpose Visual SLAM Approaches
- [논문 링크](https://arxiv.org/pdf/2107.07589.pdf)TransformerFusion: Monocular RGB Scene Reconstruction using Transformers
- [논문 링크](https://arxiv.org/pdf/2107.02191.pdf)PLF-VINS: Real-Time Monocular Visual-Inertial SLAM With Point-Line Fusion and Parallel-Line Fusion
- [논문 링크](https://ieeexplore.ieee.org/abstract/document/9478195)Architectures for SLAM and Augmented Reality Computing
- [논문 링크](https://oramavr.com/ORamaVRPublications/FPL2021_vipGPU.pdf)OdoViz: A 3D Odometry Visualization and Processing Tool
- [논문 링크](https://arxiv.org/pdf/2107.07557.pdf)Dense point cloud map construction based on stereo VINS for mobile vehicles
- [논문 링크](https://www.sciencedirect.com/science/article/abs/pii/S0924271621001672)Semantic and edge-based visual odometry by joint minimizing semantic and edge distance error
- [논문 링크](https://www.sciencedirect.com/science/article/abs/pii/S0262885621001451)Coarse-to-fine Semantic Localization with HD Map for Autonomous Driving in Structural Scenes
- [논문 링크](https://arxiv.org/pdf/2107.02557.pdf)HybVIO: Pushing the Limits of Real-time Visual-inertial Odometry
- [논문 링크](https://arxiv.org/pdf/2106.11857.pdf)A Tutorial: Mobile Robotics, SLAM, Bayesian Filter, Keyframe Bundle Adjustment and ROS Applications
- [논문 링크](https://books.google.co.kr/books?id=QQE5EAAAQBAJ&pg=PA227&lpg=PA227&dq=A+Tutorial:+Mobile+Robotics,+SLAM,+Bayesian+Filter,+Keyframe+Bundle+Adjustment+and+ROS+Applications&source=bl&ots=lNKAssOUC8&sig=ACfU3U1m_2CREi7amLbWbGcaVfR6cna2lQ&hl=ko&sa=X&ved=2ahUKEwih75jQiYXyAhVIMd4KHca8Cd8Q6AEwDXoECAIQAw#v=onepage&q=A%20Tutorial%3A%20Mobile%20Robotics%2C%20SLAM%2C%20Bayesian%20Filter%2C%20Keyframe%20Bundle%20Adjustment%20and%20ROS%20Applications&f=false)Challenges of the Application of Front-Wheel Odometry for Vehicle Localization
- [논문 링크](https://ieeexplore.ieee.org/abstract/document/9480228)Moving Object Tracking for SLAM-based Augmented Reality
- [논문 링크](https://www.researchgate.net/profile/Rodrigo-Silva-20/publication/352664821_Moving_Object_Tracking_for_SLAM-based_Augmented_Reality/links/60d2183a45851566d5837def/Moving-Object-Tracking-for-SLAM-based-Augmented-Reality.pdf)Geometry-Constrained Scale Estimation for Monocular Visual Odometry
- [논문 링크](https://ieeexplore.ieee.org/abstract/document/9479774)DT-SLAM: Dynamic Thresholding Based Corner Point Extraction in SLAM System
- [논문 링크](https://ieeexplore.ieee.org/stamp/stamp.jsp?arnumber=9464347)RAM-VO: Less is more in Visual Odometry
- [논문 링크](https://arxiv.org/pdf/2107.02974.pdf)Image Enhancement using GANs for Monocular Visual Odometry
- [논문 링크](https://ieeexplore.ieee.org/abstract/document/9468831)Dense point cloud map construction based on stereo VINS for mobile vehicles
- [논문 링크](https://www.sciencedirect.com/science/article/abs/pii/S0924271621001672)deep learning for VO or VSLAM
- [논문 링크](https://blog.katastros.com/a?ID=00600-33772170-a5ff-4031-9a54-e5cc04ef069c)Feature-Based SLAM: Why Simultaneous Localisation and Mapping?
- [논문 링크](http://www.roboticsproceedings.org/rss17/p009.pdf)Application of machine learning in SLAM algorithms
- [논문 링크](https://www.degruyter.com/document/doi/10.1515/9783110702514-009/html)Dynamic RGB-D SLAM Based on Static Probability and Observation Number
- [논문 링크](https://ieeexplore.ieee.org/abstract/document/9459604)DVT-SLAM: Deep-Learning Based Visible and Thermal Fusion SLAM
- [논문 링크](https://link.springer.com/chapter/10.1007/978-981-16-3142-9_37)Global-Map-Registered Local Visual Odometry Using On-the-Fly Pose Graph Updates
- [논문 링크](https://www.researchgate.net/profile/Masahiro-Yamaguchi-7/publication/343984706_Global-Map-Registered_Local_Visual_Odometry_Using_On-the-Fly_Pose_Graph_Updates/links/6075402692851cb4a9d82d8a/Global-Map-Registered-Local-Visual-Odometry-Using-On-the-Fly-Pose-Graph-Updates.pdf)A survey on indoor 3D modeling and applications via RGB-D devices
- [논문 링크](https://link.springer.com/article/10.1631/FITEE.2000097)Robust and Accurate RGB-D Reconstruction With Line Feature Constraints
- [논문 링크](https://ieeexplore.ieee.org/abstract/document/9468706)Collaborative Visual Inertial SLAM for Multiple Smart Phones
- [논문 링크](https://arxiv.org/pdf/2106.12186.pdf)
Implicit representations
Correspondence-Free Point Cloud Registration with SO (3)-Equivariant Implicit Shape Representations
- [논문 링크](https://arxiv.org/pdf/2107.10296.pdf)Advances in neural rendering
- [논문 링크](https://dl.acm.org/doi/abs/10.1145/3450508.3464573)Animatable Neural Radiance Fields from Monocular RGB Video
- [논문 링크](https://arxiv.org/pdf/2106.13629.pdf)
Backend (optimisation)
Consensus-Informed Optimization Over Mixtures for Ambiguity-Aware Object SLAM
- [논문 링크](https://arxiv.org/pdf/2107.09265.pdf)Continuous Integration over SO(3) for IMU Preintegration
- [논문 링크](http://www.roboticsproceedings.org/rss17/p078.pdf)ROBUST ESTIMATION IN ROBOT VISION AND PHOTOGRAMMETRY: A NEW MODEL AND ITS APPLICATIONS
- [논문 링크](https://web.archive.org/web/20210618092206id_/https://www.isprs-ann-photogramm-remote-sens-spatial-inf-sci.net/V-1-2021/137/2021/isprs-annals-V-1-2021-137-2021.pdf)Modular Multi-Sensor Fusion: A Collaborative State Estimation Perspective
- [논문 링크](https://ieeexplore.ieee.org/abstract/document/9479759)Fast and Memory Efficient Graph Optimization via ICM for Visual Place Recognition
- [논문 링크](http://www.roboticsproceedings.org/rss17/p091.pdf)Globally Optimal Consensus Maximization for Relative Pose Estimation With Known Gravity Direction
- [논문 링크](https://ieeexplore.ieee.org/abstract/document/9447984)
Front-end (Local/global features, closed-form solutions)
HDPL: a hybrid descriptor for points and lines based on graph neural networks
- [논문 링크](https://www.emerald.com/insight/content/doi/10.1108/IR-02-2021-0042/full/html)SSC: Semantic Scan Context for Large-Scale Place Recognition
- [논문 링크](https://arxiv.org/pdf/2107.00382.pdf)ESA-VLAD: A Lightweight Network Based on Second-Order Attention and NetVLAD for Loop Closure Detection
- [논문 링크](https://ieeexplore.ieee.org/abstract/document/9472966)A survey: which features are required for dynamic visual simultaneous localization and mapping?
- [논문 링크](https://vciba.springeropen.com/articles/10.1186/s42492-021-00086-w)What makes visual place recognition easy or hard?
- [논문 링크](https://arxiv.org/pdf/2106.12671.pdf)A Flexible and Efficient Loop Closure Detection Based on Motion Knowledge
- [논문 링크](https://www.researchgate.net/profile/Bingxi_Liu2/publication/353071375_A_Flexible_and_Efficient_Loop_Closure_Detection_Based_on_Motion_Knowledge/links/60e69eee0fbf460db8ede64e/A-Flexible-and-Efficient-Loop-Closure-Detection-Based-on-Motion-Knowledge.pdf)Vector Semantic Representations as Descriptors for Visual Place Recognition
- [논문 링크](http://www.roboticsproceedings.org/rss17/p083.pdf)Feature-Level Collaboration: Joint Unsupervised Learning of Optical Flow, Stereo Depth and Camera Motion
- [논문 링크](https://openaccess.thecvf.com/content/CVPR2021/papers/Chi_Feature-Level_Collaboration_Joint_Unsupervised_Learning_of_Optical_Flow_Stereo_Depth_CVPR_2021_paper.pdf)Relative scale estimation approach for monocular visual odometry
- [논문 링크](https://research.latinxinai.org/papers/cvpr/2021/pdf/59_CameraReady_59.pdf)Detection of loop closure in visual SLAM: a stacked assorted auto-encoder based approach
- [논문 링크](https://link.springer.com/article/10.1007/s11801-021-0156-9)Towards real-time monocular depth estimation for mobile systems
- [논문 링크](https://www.spiedigitallibrary.org/conference-proceedings-of-spie/11785/117850J/Towards-real-time-monocular-depth-estimation-for-mobile-systems/10.1117/12.2596031.short?SSO=1)SeqNetVLAD vs PointNetVLAD: Image Sequence vs 3D Point Clouds for Day-Night Place Recognition
- [논문 링크](https://arxiv.org/pdf/2106.11481.pdf)InFlow: Robust outlier detection utilizing Normalizing Flows
- [논문 링크](https://arxiv.org/pdf/2106.12894.pdf)Utilization of Semantic Planes: Improved Localization and Dense Semantic Map for Monocular SLAM in Urban Environment
- [논문 링크](https://ieeexplore.ieee.org/abstract/document/9462437)
3D detection / 6D Pose estimation
Visual-Inertial-Semantic Scene Representation for 3D Object Detection
- [논문 링크](https://openaccess.thecvf.com/content_cvpr_2017/papers/Dong_Visual-Inertial-Semantic_Scene_Representation_CVPR_2017_paper.pdf)Monocular 3D Object Detection: An Extrinsic Parameter Free Approach
- [논문 링크](https://openaccess.thecvf.com/content/CVPR2021/papers/Zhou_Monocular_3D_Object_Detection_An_Extrinsic_Parameter_Free_Approach_CVPR_2021_paper.pdf)6D Object Pose Estimation using Keypoints and Part Affinity Fields
- [논문 링크](https://arxiv.org/pdf/2107.02057.pdf)
Hardware / acceleration
Multiple Master-Slave FPGA Architecture of a Stereo Visual Odometry
- [논문 링크](https://ieeexplore.ieee.org/abstract/document/9492095)Related Posts