Awesome-Learning-based-VO-VIO
Paper Survey for Learning-based Odometry
https://github.com/KwanWaiPang/Awesome-Learning-based-VO-VIO
Last synced: 1 day ago
JSON representation
-
Learning-based LiDAR Odometry
- MA-SLAM: Active SLAM in Large-Scale Unknown Environment using Map Aware Deep Reinforcement Learning - --|---|
- A Generative Hierarchical Optimization Framework for LiDAR Odometry Using Conditional Diffusion Models - --|---|
- DiffLO: Semantic-Aware LiDAR Odometry with Diffusion-Based Refinement - --|
- Blog
- LiDAR Inertial Odometry And Mapping Using Learned Registration-Relevant Features - autonomy/FeatureLIOM.svg)](https://github.com/neu-autonomy/FeatureLIOM)|learning-based select points|
- LiDAR-OdomNet: LiDAR Odometry Network Using Feature Fusion Based on Attention - OdomNet.svg)](https://github.com/ParvezAlam123/LiDAR-OdomNet)|---|
- PIN-SLAM: LiDAR SLAM Using a Point-Based Implicit Neural Representation for Achieving Global Map Consistency - --|
- DeepPointMap: Advancing LiDAR SLAM with Unified Neural Descriptors - --|
- DELO: Deep Evidential LiDAR Odometry using Partial Optimal Transport
- NeRF-LOAM: Neural Implicit Representation for Large-Scale Incremental LiDAR Odometry and Mapping - LOAM.svg)](https://github.com/JunyuanDeng/NeRF-LOAM)|---|
- Translo: A window-based masked point transformer framework for large-scale lidar odometry - --|
- LONER: LiDAR Only Neural Representations for Real-Time SLAM - --|
- HPPLO-Net: Unsupervised LiDAR Odometry Using a Hierarchical Point-to-Plane Solver - Net.svg)](https://github.com/IMRL/HPPLO-Net)|---|
- Efficient 3D Deep LiDAR Odometry - Net.svg)](https://github.com/IRMVLab/EfficientLO-Net)|---|
- PWCLO-Net: Deep LiDAR Odometry in 3D Point Clouds Using Hierarchical Embedding Mask Optimization - --|
- Deeplio: deep lidar inertial sensor fusion for odometry estimation - --|
- Lodonet: A deep neural network with 2d keypoint matching for 3d lidar odometry estimation - --|---|
- Unsupervised geometry-aware deep lidar odometry - --|[website](https://sites.google.com/view/deeplo)|
- DMLO: Deep Matching LiDAR Odometry - --|---|
- Deeppco: End-to-end point cloud odometry through deep parallel neural network - --|---|
- Lo-net: Deep real-time lidar odometry - --|---|
- L3-net: Towards learning based lidar localization for autonomous driving - --|---|
- CNN for IMU assisted odometry estimation using velodyne LiDAR - --|---|
- Deep learning for laser based odometry estimation - --|---|
- LIR-LIVO: A Lightweight, Robust LiDAR/Vision/Inertial Odometry with Illumination-Resilient Deep Features - A-CAT/LIR-LIVO.svg)](https://github.com/IF-A-CAT/LIR-LIVO)|Fast-LIVO+AirSLAM|
- LiDAR-Visual-Inertial Tightly-coupled Odometry with Adaptive Learnable Fusion Weights - --|F-LOAM+VINS-Mono|
- |RAMP-VO|
- Leap-vo: Long-term effective any point tracking for visual odometry
- Efficient Camera Exposure Control for Visual Odometry via Deep Reinforcement Learning - --|
- Reinforcement learning meets visual odometry - rpg/rl_vo.svg)](https://github.com/uzh-rpg/rl_vo)|[Blog](https://kwanwaipang.github.io/RL-for-VO/)|
- Deep event visual odometry - vision/DEVO.svg)](https://github.com/tum-vision/DEVO)|---|
- Multi-Session SLAM with Differentiable Wide-Baseline Pose Optimization - vl/MultiSlam_DiffPose.svg)](https://github.com/princeton-vl/MultiSlam_DiffPose)|Multi VO|
- Deep patch visual slam - vl/DPVO.svg)](https://github.com/princeton-vl/DPVO)|---|
- Deep patch visual odometry - vl/DPVO.svg)](https://github.com/princeton-vl/DPVO)|---|
- Dytanvo: Joint refinement of visual odometry and motion segmentation in dynamic environments - --|
- Ganvo: Unsupervised deep monocular visual odometry and depth estimation with generative adversarial networks - --|---|
- Tartanvo: A generalizable learning-based vo - --|
- DROID-SLAM: Deep Visual SLAM for Monocular,Stereo, and RGB-D Cameras - vl/DROID-SLAM.svg)](https://github.com/princeton-vl/DROID-SLAM)|---|
- Generalizing to the open world: Deep visual odometry with online adaptation - --|---|
- DF-VO: What should be learnt for visual odometry? - Zhan/DF-VO.svg)](https://github.com/Huangying-Zhan/DF-VO)|---|
- Visual odometry revisited: What should be learnt? - Zhan/DF-VO.svg)](https://github.com/Huangying-Zhan/DF-VO)|---|
- Towards better generalization: Joint depth-pose learning without posenet - --|
- Diffposenet: Direct differentiable camera pose estimation - --|---|
- D3vo: Deep depth, deep pose and deep uncertainty for monocular visual odometry - --|[website](https://cvg.cit.tum.de/research/vslam/d3vo)|
- Unsupervised scale-consistent depth learning from video - SfMLearner-Release.svg)](https://github.com/JiawangBian/SC-SfMLearner-Release)<br>[](https://github.com/JiawangBian/sc_depth_pl)|---|
- Unsupervised scale-consistent depth and ego-motion learning from monocular video - SfMLearner-Release.svg)](https://github.com/JiawangBian/SC-SfMLearner-Release)<br>[](https://github.com/JiawangBian/sc_depth_pl)|---|
- Recurrent neural network for (un-) supervised learning of monocular video visual odometry and depth - |-|
- Pose graph optimization for unsupervised monocular visual odometry - --|---|
- Raum-vo: Rotational adjusted unsupervised monocular visual odometry - --|---|
- DeepAVO: Efficient pose refining with feature distilling for deep Visual Odometry - --|---|
- Undeepvo: Monocular visual odometry through unsupervised deep learning - |[website](https://senwang.gitlab.io/UnDeepVO/)|
- Unsupervised learning of monocular depth estimation and visual odometry with deep feature reconstruction - Zhan/Depth-VO-Feat.svg)](https://github.com/Huangying-Zhan/Depth-VO-Feat)|---|
- Deepvo: Towards end-to-end visual odometry with deep recurrent convolutional neural networks - |[website](https://senwang.gitlab.io/DeepVO/)|
- Unsupervised learning of depth and ego-motion from video - --|
- Dh-ptam: a deep hybrid stereo events-frames parallel tracking and mapping system - PTAM.svg)](https://github.com/AbanobSoliman/DH-PTAM)|Superpoint+[stereo ptam](https://github.com/uoip/stereo_ptam)|
- Scene-agnostic Pose Regression for Visual Localization - based|
- Image as an IMU: Estimating Camera Motion from a Single Motion-Blurred Image - --|Velometer|
- Online Adaptive Keypoint Extraction for Visual Odometry Across Different Scenes - --|Reinforcement Learning|
- MAC-VO: Metrics-aware Covariance for Learning-based Stereo Visual Odometry - VO/MAC-VO.svg)](https://github.com/MAC-VO/MAC-VO)|[website](https://mac-vo.github.io/)|
- LoSeVO: Local Sequence Constraints for Deep Visual Odometry - --|---|
- VOCAL: Visual Odometry via ContrAstive Learning - --|---|
- Continual Learning of Regions for Efficient Robot Localization on Large Maps - BioLab/continual-learning-regions.svg)](https://github.com/MI-BioLab/continual-learning-regions)|---|
- Occlusion-Aware Monocular Visual Odometry for Robust Trajectory Tracking - --|---|
- CoProU-VO: Combining Projected Uncertainty for End-to-End Unsupervised Monocular Visual Odometry - Xie/CoProU.svg)](https://github.com/Jchao-Xie/CoProU)|[website](https://jchao-xie.github.io/CoProU/)|
- Lp-slam: language-perceptive RGB-D SLAM framework exploiting large language model - --|
- Policies over Poses: Reinforcement Learning based Distributed Pose-Graph Optimization for Multi-Robot SLAM - uga/policies-over-poses.svg)](https://github.com/herolab-uga/policies-over-poses)|---|
- Lp-slam: language-perceptive RGB-D SLAM framework exploiting large language model - --|
- Lp-slam: language-perceptive RGB-D SLAM framework exploiting large language model - --|
- Lp-slam: language-perceptive RGB-D SLAM framework exploiting large language model - --|
- Lp-slam: language-perceptive RGB-D SLAM framework exploiting large language model - --|
- Real-Time Indoor Object SLAM with LLM-Enhanced Priors - --|---|
- A Robust and Accurate Stereo SLAM Based on Learned Feature Extraction and Matching - --|---|
- Lp-slam: language-perceptive RGB-D SLAM framework exploiting large language model - --|
- Lp-slam: language-perceptive RGB-D SLAM framework exploiting large language model - --|
- Lp-slam: language-perceptive RGB-D SLAM framework exploiting large language model - --|
- Lp-slam: language-perceptive RGB-D SLAM framework exploiting large language model - --|
-
Other Related Resource
- 3dfeat-net: Weakly supervised local 3d features for point cloud registration - --|
- Blog
- Paper List
- A Consistency-Aware Spot-Guided Transformer for Versatile and Hierarchical Point Cloud Registration - --|
- 3dregnet: A deep neural network for 3d point registration - --|
- P2b: Point-to-box network for 3d object tracking in point clouds - --|---|
- Pointconv: Deep convolutional networks on 3d point clouds - --|---|
- Meteornet: Deep learning on dynamic 3d point cloud sequences - --|---|
- Deep closest point: Learning representations for point cloud registration - --|---|
- Pointnetlk: Robust & efficient point cloud registration using pointnet - --|
- 3D local features for direct pairwise registration - --|---|
- Kpconv: Flexible and deformable convolution for point clouds - --|
- Deepvcp: An end-to-end deep neural network for point cloud registration - --|---|
- Flownet3d: Learning scene flow in 3d point clouds - --|---|
- Rangenet++: Fast and accurate lidar semantic segmentation - --|---|
- Dynamic graph cnn for learning on point clouds - --|---|
- Paper List
- Blog
- Blog
- To Match or Not to Match: Revisiting Image Matching for Reliable Visual Place Recognition - --|Matching|
- Learning Affine Correspondences by Integrating Geometric Constraints
- Paper List - ->
- Blog - ->
- Blog - ->
- Blog - ->
- LiftFeat: 3D Geometry-Aware Local Feature Matching - deeplearning/LiftFeat.svg)](https://github.com/lyp-deeplearning/LiftFeat)|---|
- EdgePoint2: Compact Descriptors for Superior Efficiency and Accuracy - --|
- Projective manifold gradient layer for deep rotation regression
- Mvsnet: Depth inference for unstructured multi-view stereo - --|---|
- Superpoint: Self-supervised interest point detection and description - yyjau/pytorch-superpoint)<br>[event superpoint](https://github.com/mingyip/pytorch-superpoint)|
- Paper List
- Theseus: A library for differentiable nonlinear optimization - ai/)|
- Superglue: Learning feature matching with graph neural networks - --|
- Tartanair: A dataset to push the limits of visual slam - dataset/)|
- Raft: Recurrent all-pairs field transforms for optical flow - vl/RAFT.svg)](https://github.com/princeton-vl/RAFT)|---|
- Metric3d v2: A versatile monocular geometric foundation model for zero-shot metric depth and surface normal estimation
- Flownet: Learning optical flow with convolutional networks - --|---|
- DeDoDe: Detect, don't describe—Describe, don't detect for local feature matching
- slides
- DeDoDe v2: Analyzing and Improving the DeDoDe Keypoint Detector - pytorch-inference)|
- Stereoglue: Robust estimation with single-point solvers
- RAFT-Stereo: Multilevel Recurrent Field Transforms for Stereo Matching - vl/RAFT-Stereo.svg)](https://github.com/princeton-vl/RAFT-Stereo)|---|
- Selecting and Pruning: A Differentiable Causal Sequentialized State-Space Model for Two-View Correspondence Learning - --|Mamba two-view correspondence|
- SupeRANSAC: One RANSAC to Rule Them All - --|
- MP-SfM: Monocular Surface Priors for Robust Structure-from-Motion - --|
- Structure-from-motion revisited
- Voxnet: A 3d convolutional neural network for real-time object recognition - --|---|
- TurboReg: TurboClique for Robust and Efficient Point Cloud Registration - 3DV/TurboReg.svg)](https://github.com/Laka-3DV/TurboReg)|---|
- Point transformer v3: Simpler faster stronger - --|
- P2b: Point-to-box network for 3d object tracking in point clouds - --|---|
- Pointnet++: Deep hierarchical feature learning on point sets in a metric space - --|---|
- Pointnet: Deep learning on point sets for 3d classification and segmentation - --|---|
- Voxnet: A 3d convolutional neural network for real-time object recognition - --|---|
-
Learning-based VIO
- SuperEvent: Cross-Modal Learning of Event-based Keypoint Detection
- CUAHN-VIO: Content-and-uncertainty-aware homography network for visual-inertial odometry - VIO.svg)](https://github.com/tudelft/CUAHN-VIO)|---|
- Airslam: An efficient and illumination-robust point-line visual slam system - lab/AirSLAM.svg)](https://github.com/sair-lab/AirSLAM)|---|
- SF-Loc: A Visual Mapping and Geo-Localization System based on Sparse Visual Structure Frames - WHU/SF-Loc.svg)](https://github.com/GREAT-WHU/SF-Loc)|[test](https://kwanwaipang.github.io/SF-Loc/)|
- SelfVIO: Self-supervised deep monocular Visual–Inertial Odometry and depth estimation - --|---|
- DEIO: Deep Event Inertial Odometry - hku/DEIO.svg)](https://github.com/arclab-hku/DEIO)|[website](https://kwanwaipang.github.io/DEIO/)|
- Adaptive vio: Deep visual-inertial odometry with online continual learning - --|---|
- DBA-Fusion: Tightly Integrating Deep Dense Visual Bundle Adjustment with Multiple Sensors for Large-Scale Localization and Mapping - WHU/DBA-Fusion.svg)](https://github.com/GREAT-WHU/DBA-Fusion)|---|
- DVI-SLAM: A dual visual inertial SLAM network - --|---|
- Bamf-slam: bundle adjusted multi-fisheye visual-inertial slam using recurrent field transforms - --|---|
- Efficient deep visual and inertial odometry with adaptive visual modality selection - Selective-VIO.svg)](https://github.com/mingyuyng/Visual-Selective-VIO)|---|
- A self-supervised, differentiable Kalman filter for uncertainty-aware visual-inertial odometry - --|---|
- Unsupervised monocular visual-inertial odometry network - --|
- Unsupervised deep visual-inertial odometry with online error correction for RGB-D imagery - --|---|
- Selective sensor fusion for neural visual-inertial odometry - --|---|
- Deepvio: Self-supervised deep learning of monocular visual inertial odometry using 3d geometric constraints - --|---|
- Vinet: Visual-inertial odometry as a sequence-to-sequence learning problem - official implementation|
- Blog
- SuperPoint-SLAM3: Augmenting ORB-SLAM3 with Deep Features, Adaptive NMS, and Learning-Based Loop Closure - --|
- OKVIS2-X: Open Keyframe-based Visual-Inertial SLAM Configurable with Dense Depth or LiDAR, and GNSS - mrl/OKVIS2-X.svg)](https://github.com/ethz-mrl/OKVIS2-X)|---|
- LightVO: Lightweight inertial-assisted monocular visual odometry with dense neural networks - --|---|
- SuperEIO: Self-Supervised Event Feature Learning for Event Inertial Odometry - hku/SuperEIO.svg)](https://github.com/arclab-hku/SuperEIO)|[website](https://arclab-hku.github.io/SuperEIO/)|
-
Learning-based Inertial Odometry
- End-to-end deep learning framework for real-time inertial attitude estimation using 6dof imu - --|---|
- AirIMU: Learning uncertainty propagation for inertial odometry
- Learned inertial odometry for autonomous drone racing - rpg/learned_inertial_model_odometry.svg)](https://github.com/uzh-rpg/learned_inertial_model_odometry)|---|
- AirIO: Learning Inertial Odometry with Enhanced IMU Feature Observability - IO/Air-IO.svg)](https://github.com/Air-IO/Air-IO)|[website](https://air-io.github.io/)<br>[Test](https://kwanwaipang.github.io/AirIO/)|
- Deep learning for inertial positioning: A survey - --|---|
- Blog
- Improved state propagation through AI-based pre-processing and down-sampling of high-speed inertial data - --|---|
- Deep IMU Bias Inference for Robust Visual-Inertial Odometry with Factor Graphs - --|---|
- IMU Data Processing For Inertial Aided Navigation:A Recurrent Neural Network Based Approach - --|---|
- Tlio: Tight learned inertial odometry
- AI-IMU dead-reckoning - imu-dr.svg)](https://github.com/mbrossar/ai-imu-dr)|---|
