Awesome CV Works · RealCat

The post contains papers-with-code about SLAM, Pose/Object tracking, Depth/Disparity/Flow Estimation, 3D-graphic, Machine Learning, Deep Learning etc.

Here is the repo: Recent-Stars-2022

I posted the content of the repo as follows.

[SLAM][ORB-SLAM3: An Accurate Open-Source Library for Visual, Visual-Inertial and Multi-Map SLAM](https://github.com/UZ-SLAMLab/ORB\_SLAM3), [PDF]
[SLAM][LIO-SAM](https://github.com/TixiaoShan/LIO-SAM)，激光雷达 IMU 紧耦合 SLAM
[Tool][Robotics Toolbox for Python](https://github.com/petercorke/robotics-toolbox-python), a Python implementation of the Robotics Toolbox for MATLAB®
[Matching][LISRD](https://github.com/rpautrat/LISRD), ECCV 2020, [PDF]，在线局部不变特征匹配！重要！
[Matching][AdaLAM](https://github.com/cavalli1234/AdaLAM)，特征匹配快速滤除外点
[Calib][fisheye_pinhole_calib_demo](https://github.com/3DCVer/fisheye\_pinhole\_calib\_demo)，包括鱼眼模型、针孔模型的相机标定，封装了自动编译、库的打包以及外部库的调用测试
[Calib][SensorCalibration](https://github.com/FENGChenxi0823/SensorCalibration), IMU 雷达标定
[VO][Low-Drift Visual Odometry in Structured Environments by Decoupling Rotational and Translational Motion](https://github.com/PyojinKim/LPVO), ICRA 2018, [PDF]，结构化环境中将旋转量与平移量进行分离优化
[VIO][VIO-SLAM](https://github.com/iamwangyabin/VIO-SLAM)，从零开始手写 VIO 课后作业
[Matching][TFMatch: Learning-based image matching in TensorFlow](https://github.com/lzx551402/tfmatch), TensorFlow 实现的 GeoDesc, ASLFeat 以及 ContextDesc
[Tutorial][SLAM-BOOK](https://github.com/yanyan-li/SLAM-BOOK)，一本关于 SLAM 的书稿，清楚的介绍 SLAM 系统中的使用的几何方法和深度学习方法，持续更新中
[Loop Closing][OverlapNet - Loop Closing for 3D LiDAR-based SLAM](https://github.com/PRBonn/OverlapNet), RSS 2020, [PDF], 3D 激光雷达 SLAM 闭环
[SLAM][VDO_SLAM](https://github.com/halajun/VDO\_SLAM), RGB-D 相机数据作为输入，实现追踪动态物体 SLAM 的功能，[PDF]
[SLAM][orbslam-map-saving-extension](https://github.com/TUMFTM/orbslam-map-saving-extension)，在 ORB-SLAM 的基础上增加保存+加载地图功能
[Tutorial][Modern Robotics: Mechanics, Planning, and Control Code Library](https://github.com/NxRLab/ModernRobotics)，现代机器人学，**\[[Homepage](http://hades.mech.northwestern.edu/index.php/Modern_Robotics)\]**
[Matching][image-matching-benchmark-baselines](https://github.com/vcg-uvic/image-matching-benchmark-baselines)，图像特征匹配挑战赛主页
[Matching][GraphLineMatching](https://github.com/mameng1/GraphLineMatching)
[Matching][Locality Preserving Matching](https://github.com/jiayi-ma/LPM), IJCAI 2017, [PDF]
[IMU][IMUOrientationEstimator](https://github.com/ydsf16/IMUOrientationEstimator)
[Feature][BEBLID: Boosted Efficient Binary Local Image Descriptor](https://github.com/iago-suarez/BEBLID)
[Relocalization][KFNet: Learning Temporal Camera Relocalization using Kalman Filtering](https://github.com/zlthinker/KFNet), CVPR 2020, [PDF]
[Matching][image-matching-benchmark](https://github.com/vcg-uvic/image-matching-benchmark)
[Matching][GMS: Grid-based Motion Statistics for Fast, Ultra-robust Feature Correspondence](https://github.com/JiawangBian/GMS-Feature-Matcher), CVPR 17 & IJCV 19, [PDF], [Project page]
[Reloc][GN-Net-Benchmark](https://github.com/Artisense-ai/GN-Net-Benchmark), CVPR 2020, GN-Net: The Gauss-Newton Loss for Multi-Weather Relocalization, [PDF], [Project page]
[Matching][SuperGluePretrainedNetwork](https://github.com/magicleap/SuperGluePretrainedNetwork), CVPR 2020, [PDF]，划重点！2020 年 sota 超大视角 2D 特征匹配，Blog
[Feature][D3Feat](https://github.com/XuyangBai/D3Feat), CVPR 2020, [PDF]
[Feature][ASLFeat](https://github.com/lzx551402/ASLFeat), CVPR 2020, ASLFeat: Learning Local Features of Accurate Shape and Localization, [PDF]
[Feature][3DFeatNet](https://github.com/yewzijian/3DFeatNet), ECCV 2018, [PDF]
[Tutorial][AutonomousDrivingCookbook](https://github.com/microsoft/AutonomousDrivingCookbook)，Scenarios, tutorials and demos for Autonomous Driving
[Tutorial][SLAMPaperReading](https://github.com/PaoPaoRobot/SLAMPaperReading)，泡泡机器人北京线下 SLAM 论文分享资料
[Tutorial][VIO_Tutotial_Course](https://github.com/lishuwei0424/VIO\_Tutotial\_Course)
[Tutorial][VO-SLAM-Review](https://github.com/MichaelBeechan/VO-SLAM-Review)
[Tutorial][VINS-Mono-code-annotation](https://github.com/QingSimon/VINS-Mono-code-annotation), VINS-Mono 代码注释以及公式推导
[Tutorial][VINS-Mono-Learning](https://github.com/ManiiXu/VINS-Mono-Learning), VINS-Mono 代码注释
[Tutorial][VINS-Course](https://github.com/HeYijia/VINS-Course), VINS-Mono code without Ceres or ROS
[Tutorial][VIO-Doc](https://github.com/StevenCui/VIO-Doc)，主流 VIO 论文推导及代码解析
[VO][CNN-DSO](https://github.com/muskie82/CNN-DSO), Direct Sparse Odometry with CNN Depth Prediction
[VO][fisheye-ORB-SLAM](https://github.com/lsyads/fisheye-ORB-SLAM), A real-time robust monocular visual SLAM system based on ORB-SLAM for fisheye cameras, without rectifying or cropping the input images
[VO][ORB_Line_SLAM](https://github.com/robotseu/ORB\_Line\_SLAM), Real-Time SLAM with BoPLW Pairs for Stereo Cameras, with Loop Detection and Relocalization Capabilities
[VO][DeepVO-pytorch](https://github.com/ChiWeiHsiao/DeepVO-pytorch.git), ICRA 2017 DeepVO: Towards end-to-end visual odometry with deep Recurrent Convolutional Neural Networks
[Calib][CamOdomCalibraTool](https://github.com/MegviiRobot/CamOdomCalibraTool), The tool to calibrate extrinsic param between camera and wheel.
[Calib][lidar_camera_calibration](https://github.com/heethesh/lidar\_camera\_calibration), another version
[Calib][OdomLaserCalibraTool](https://github.com/MegviiRobot/OdomLaserCalibraTool.git)，相机与 2D 雷达标定
[Calib][extrinsic_lidar_camera_calibration](https://github.com/UMich-BipedLab/extrinsic\_lidar\_camera\_calibration), LiDARTag: A Real-Time Fiducial Tag using Point Clouds, arXiv 2019, [PDF]
[Calib][velo2cam_calibration](https://github.com/beltransen/velo2cam\_calibration), Automatic Calibration algorithm for Lidar-Stereo camera, [Project page]
[Dataset][IRS: A Large Synthetic Indoor Robotics Stereo Dataset for Disparity and Surface Normal Estimation](https://github.com/HKBU-HPML/IRS.git)
[Tools][averaging-quaternions](https://github.com/christophhagen/averaging-quaternions)，四元数平均

分割线，以下是 2019 年的星标项目，上面是 2020 年新星标的。

R2D2: Reliable and Repeatable Detector and Descriptor, NeurIPS 2019, [PDF], [Project page]，深度学习特征点+描述子
Semantic_SLAM，语义 SLAM：ROS + ORB SLAM + PSPNet101
PlaceRecognition-LoopDetection, Light-weight place recognition and loop detection using road markings
DOOR-SLAM: Distributed, online, and outlier resilient SLAM for robotic teams, [PDF], [Project page]，多机器人协作 SLAM，增强了场景的适用性
awesome-local-global-descriptor，超详细深度学习特征点描述子集合，需要重点关注一下这个 repo
GIFT: Learning Transformation-Invariant Dense Visual Descriptors via Group CNNs, NeurIPS 2019，[PDF], [Project page]，浙大 CAD+商汤联合实验室出品，利用 Group CNN 来改进 superpoint 描述子（仅描述，特征点提取可任意选择），可以大幅度增强视角变化时的特征点复检率与匹配点数
Key.Net: Keypoint Detection by Handcrafted and Learned CNN Filters, ICCV 2019, [PDF]，深度学习特征点
Self-Supervised 3D Keypoint Learning for Ego-motion Estimation, [PDF], [YouTube]，深度学习特征点
VINS-Mono-Optimization，实现点线紧耦合优化的 VINS-Mono
msckf_vio 注释版本
NetVLAD-pytorch, NetVLAD 场景识别的 PyTorch 实现
High-Precision Localization Using Ground Texture (Micro-GPS), ECCV 2018, [PDF], [Project page], [code]，地向（摄像机朝向地面）SLAM，获得高精度重定位效果。
PlaneSLAM, Paper: “On the Representation of Planes for Efficient Graph-based SLAM with High-level Features”
XIVO: X Inertial-aided Visual Odometry and Sparse Mapping, an open-source repository for visual-inertial odometry/mapping.
DeepTAM, ECCV 2018, [PDF], [Project page], a learnt system for keyframe-based dense camera tracking and mapping.
iRotAvg, Why bundle adjust?, ICRA 2019, [PDF]
Tactical Rewind: Self-Correction via Backtracking in Vision-and-Language Navigation, CVPR 2019, [PDF]，视觉+语言导航
DOOR-SLAM
An Evaluation of Feature Matchers for Fundamental Matrix Estimation, BMVC 2019, [PDF], [Project Page]，特征匹配
A Tightly Coupled 3D Lidar and Inertial Odometry and Mapping Approach, ICRA 2019, [PDF], [Project Page]，紧耦合雷达+IMU SLAM
On the Representation of Planes for Efficient Graph-based SLAM with High-level Features，利用平面信息的 SLAM
Visual Odometry Revisited: What Should Be Learnt?, arXiv 2019, [PDF]，深度学习深度+光流进行 VO
RF-Net: An End-to-End Image Matching Network based on Receptive Field, CVPR 2019, [PDF]，端到端图像匹配
Fast-Planner, IEEE Robotics and Automation Letters (RA-L), 2019, [PDF]，无人机轨迹生成
A general and flexible factor graph non-linear least square optimization framework, CoRR 2019, [PDF], [Project Page]
Demo for Kalman filter in ranging system，卡尔曼滤波原理演示
A Holistic Visual Place Recognition Approach using Lightweight CNNs for Severe ViewPoint and Appearance Changes，场景识别（外观与视角变化时），训练和部署源码
SIPs: Succinct Interest Points from Unsupervised Inlierness Probability Learning,3D Vision (3DV) 2019, [PDF]，RPG 实验室出品，深度学习特征点（有特征描述子）
Matching Features Without Descriptors: Implicitly Matched Interest Points, BMVC 2019, [PDF], RPG 实验室出品，无需特征描述即可进行特征匹配
Learning Lightweight Lane Detection CNNs by Self Attention Distillation (ICCV 2019), ICCV 2019, [PDF]，深度学习道路检测
Awesome SLAM Datasets，史上最全 SLAM 数据集，公众号说明：最全 SLAM 开源数据集
GNSS-INS-SIM，惯导融合模拟器，支持 IMU 数据，轨迹生成等
Multi-Sensor Combined Navigation Program(GNSS, IMU, Camera and so on) 多源多传感器融合定位 GPS/INS 组合导航
SOSNet: Second Order Similarity Regularization for Local Descriptor Learning, CVPR 2019, [Project page] [Paper] [Poster] [Slides]，一种深度学习特征描述子
Look No Deeper: Recognizing Places from Opposing Viewpoints under Varying Scene Appearance using Single-View Depth Estimation, ICRA 2019, [PDF]，利用深度图像实现了大视角长时间的场景识别（根据深度图筛选得到不同深度层次的特征点然后与当前帧进行匹配，提高了场景召回率）
CALC2.0, Convolutional Autoencoder for Loop Closure 2.0，用于闭环检测
SegMap, RSS 2018, [PDF]，一种基于 3D 线段的地图表示，可用于场景识别/机器人定位/环境重建等
MSCKF_VIO, a stereo version of MSCKF，基于 MSCKF 的双目 VIO
NetVLAD: CNN architecture for weakly supervised place recognition，CVPR 2016, CNN 框架弱监督学习场景识别，[Project Page]
easy_handeye, Simple, straightforward ROS library for hand-eye calibration
SuperPoint-SLAM，利用 SuperPoint 替换 ORB 特征点
PyRobot: An Open Source Robotics Research Platform
From Coarse to Fine: Robust Hierarchical Localization at Large Scale with HF-Net, [PDF]
Super fast implementation of ICP in CUDA
A generic interface for disparity map and pointcloud insertion
SPHORB: A Fast and Robust Binary Feature on the Sphere, International Journal of Computer Vision 2015, [PDF], [Project Page]
BADSLAM: Bundle Adjusted Direct RGB-D SLAM, CVPR 2019, [PDF]
High Speed and High Dynamic Range Video with an Event Camera, arXiv, [PDF], [Project Page]
Awesome-VIO, Discuss about VIO in PaoPaoRobot group
GyroAllan，陀螺仪随机误差的 Allan 方差分析，Another version
Self-supervised Sparse-to-Dense: Self-supervised Depth Completion from LiDAR and Monocular Camera, ICRA 2019, [PDF]，优化 LiDAR 以及单目得到的深度图
PlaneRCNN: 3D Plane Detection and Reconstruction from a Single Image, CVPR 2019, [PDF], [Project Page]，通过单幅图像进行 3D 平面检测以及重建
DBow3，注释版的 DBow3 代码
Visual-Inertial Mapping with Non-Linear Factor Recovery, [PDF], [Project Page]，时空联合的 VIO 优化方案
ICRA2019-paper-list, ICRA 2019 论文列表（泡泡机器人出品暂时无链接）
Fast Cylinder and Plane Extraction from Depth Cameras for Visual Odometry, IROS 2018, [PDF]，利用深度图进行圆柱检测以及平面检测进行 VO
Solutions to assignments of Robot Mapping Course WS 2013/14 by Dr. Cyrill Stachniss at University of Freiburg, SLAM 算法学习课后作业答案
Direct sparse odometry combined with stereo cameras and IMU，双目 DSO+IMU
Direct Sparse Odometry with Stereo Cameras，双目 DSO
Python binding of SLAM graph optimization framework g2o, Python 版本的 g2o 实现
SuperPoint: Self-Supervised Interest Point Detection and Description, CVPR 2018, [Paper]，深度学习描述子+描述
ContextDesc: Local Descriptor Augmentation with Cross-Modality Context, CVPR 2019, [Paper]，深度学习描述子
D2-Net: A Trainable CNN for Joint Description and Detection of Local Features, CVPR 2019, [Paper], [Project Page]，深度学习关键点+描述
ROS interface for ORBSLAM2, ROS 版本的 ORBSLAM2
CNN-SVO: Improving the Mapping in Semi-Direct Visual Odometry Using Single-Image Depth Prediction，[Paper]
VINS-Mono-Learning，代码注释版 VINS-Mono，初学者学习
OpenVSLAM: Versatile Visual SLAM Framework, [Project Page]
RESLAM: A real-time robust edge-based SLAM system, ICRA 2019, [Paper]
PL-SLAM: a Stereo SLAM System through the Combination of Points and Line Segments, [Paper]，线特征 SLAM
Good Line Cutting: towards Accurate Pose Tracking of Line-assisted VO/VSLAM, ECCV 2018, [Project Page]，改进的 PL-SLAM
Spherical Regression: Learning Viewpoints, Surface Normals and 3D Rotations on n-Spheres, CVPR 2019, [Paper]
svo_edgelet，在线轨迹生成
Drone SLAM project for Caltech’s ME 134 Autonomy class, [PDF]
Online Trajectory Generation of a MAV for Chasing a Moving Target in 3D Dense Environments, [Paper]
PythonRobotics, [Paper], CppRobotics
Bundle adjustment demo using Ceres Solver, [Blog], ceres 实现 BA
CubeSLAM: Monocular 3D Object Detection and SLAM, [Paper]
PointRCNN: 3D Object Proposal Generation and Detection from Point Cloud, CVPR 2019, [Paper]
GIST-Global Image Descriptor, GIST 描述子
mav voxblox planning, MAV planning tools using voxblox as the map representation.
Python Kalman Filter, 30 行实现卡尔曼滤波
vicalib，视觉惯导系统标定工具
BreezySLAM，基于雷达的 SLAM，支持 Python(&MATLAB, C++, and Java)
Probabilistic-Robotics, 《概率机器人》中文版，书和课后习题
Stanford Self Driving Car Code, [Paper]，斯坦福自动驾驶车代码
Udacity Self-Driving Car Engineer Nanodegree projects
Artificial Intelligence in Automotive Technology, TUM 自动驾驶技术中的人工智能课程
DeepMatchVO: Beyond Photometric Loss for Self-Supervised Ego-Motion Estimation, ICRA 2019, [Paper]
GSLAM: A General SLAM Framework and Benchmark, CVPR 2019, [Paper]，集成了各种传感器输入的 SLAM 统一框架
Visual-Odometric Localization and Mapping for Ground Vehicles Using SE(2)-XYZ Constraints，ICRA 2019，基于 SE(2)-XYZ 约束的 VO 系统
Simple bag-of-words loop closure for visual SLAM, [Blog]，回环
FBOW (Fast Bag of Words), an extremely optimized version of the DBow2/DBow3 libraries，优化版本的 DBow2/DBow3
Multi-State Constraint Kalman Filter (MSCKF) for Vision-aided Inertial Navigation(master’s thesis)
MSCKF, MSCKF 中文注释版
Calibration algorithm for a camera odometry system, VO 系统的标定程序
Modified version of VINS-Mono，注释版本 VINS Mono
Extreme Relative Pose Estimation for RGB-D Scans via Scene Completion, [Paper]
Implementation of EPnP algorithm with Eigen，利用 Eigen 编写的 EPnP
Real-time SLAM system with deep features，深度学习描述子（ORB vs. GCNv2）
Unsupervised Learning of Monocular Depth Estimation and Visual Odometry with Deep Feature Reconstruction, CVPR 2018，无监督单目深度恢复以及 VO
ORB-SLAM-windows, Windows 版本的 ORB-SLAM
StructVIO : Visual-inertial Odometry with Structural Regularity of Man-made Environments, [Project Page]
KalmanFiltering，各种卡尔曼滤波器的 demo
Stereo Odometry based on careful Feature selection and Tracking, [Paper], C++ OpenCV 实现 SOFT
Visual SLAM with RGB-D Cameras based on Pose Graph Optimization
Multi-threaded generic RANSAC implementation，多线程 RANSAC
Visual Odometry with Drift-Free Rotation Estimation Using Indoor Scene Regularities, BMVC 2017, [Project Page]，利用平面正交信息进行 VO
ICE-BA, CVPR 2018, [Paper]
GraphSfM: Robust and Efficient Graph-based Structure from Motion, [Project Page]
LOAM_NOTED, loam 中文注解版
Divide and Conquer: Efficient Density-Based Tracking of 3D Sensors in Manhattan Worlds, ACCV 2016, [Project Page]，曼哈顿世界利用深度传感器进行旋转量平移量分离优化
Real-time Manhattan World Rotation Estimation in 3D, IROS 2015，实时曼哈顿世界旋转估计
Event-based Vision Resources，关于事件相机的资源
AutonomousVehiclePaper，无人驾驶相关论文速递
Segmentation.X, Segmentation 相关论文&代码
CVPR-2019, CVPR 2019 论文开源项目合集
awesome-slam, SLAM 合集
awesome-visual-slam，视觉 SLAM 合集
Papers with code，周更论文 with 代码
Awesome Human Pose Estimation, awesome-object-pose，位姿估计合集
MVision，大礼包：机器人视觉移动机器人 VS-SLAM ORB-SLAM2 深度学习目标检测 yolov3 行为检测 opencv PCL 机器学习无人驾驶

Pose/Object tracking

Unsupervised person re-identification by soft multilabel learning, CVPR 2019, [Paper]
FCOS: Fully Convolutional One-Stage Object Detection, ICCV 2019, [Paper]
Hand Detection and Orientation Estimation
Spatial-Temporal Person Re-identification, AAAI 2019, [Paper]
A tiny, friendly, strong pytorch implement of person re-identification baseline. Tutorial, CVPR 2019, [Paper]
Progressive Pose Attention for Person Image Generation, CVPR 2019, [Paper]
FSA-Net: Learning Fine-Grained Structure Aggregation for Head Pose Estimation from a Single Image, CVPR 2019, [Paper]
An unofficial implementation for paper “Fast Human Pose Estimation”, CVPR 2019, [Paper]
Real-time single person pose estimation for Android and iOS，手机端实现人体位姿估计
Basics of 2D and 3D Human Pose Estimation，人体姿态估计入门
Libra R-CNN: Towards Balanced Learning for Object Detection
High-resolution networks (HRNets) for object detection, [Paper]
Learning Correspondence from the Cycle-Consistency of Time, CVPR 2019, [Paper]
PVNet: Pixel-wise Voting Network for 6DoF Pose Estimation, CVPR 2019, [Paper], [Project Page]
Self-Supervised Learning of 3D Human Pose using Multi-view Geometry, CVPR 2018, [Paper]
PifPaf: Composite Fields for Human Pose Estimation, [Paper]
Deep High-Resolution Representation Learning for Human Pose Estimation, CVPR 2019, [Paper], [Project Page]
PoseFlow: Efficient Online Pose Tracking), BMVC 2018, [Paper]
A Bottom-Up Clustering Approach to Unsupervised Person Re-identification，AAAI 2019，重定位
Fast Online Object Tracking and Segmentation: A Unifying Approach, CVPR 2019, [Paper] [Video] [Project Page]
SimpleDet - A Simple and Versatile Framework for Object Detection and Instance Recognition, [Paper]

Depth/Disparity & Flow estimation

[Depth][SemiGlobalMatching](https://github.com/ethan-li-coding/SemiGlobalMatching), SGM 双目立体匹配算法完整实现，代码规范，注释丰富且清晰，CSDN 同步教学
PointMVSNet: Point-based Multi-view Stereo Network, ICCV 2019, [Paper]
DeepLiDAR, CVPR 2019, [Paper]，单张 RGB 图像+稀疏雷达数据进行室外场景深度估计
Real-Time Monocular Depth Estimation using Synthetic Data with Domain Adaptation via Image Style Transfer, CVPR 2018, [Paper]
Learning Single-Image Depth from Videos using Quality Assessment Networks, CVPR 2019, [Paper], [Project Page]
SCDA: Adapting Object Detectors via Selective Cross-Domain Alignment, CVPR 2019, [Paper], [Project Page]
Learning monocular depth estimation infusing traditional stereo knowledge, CVPR 2019, [PDF]
HPLFlowNet: Hierarchical Permutohedral Lattice FlowNet for Scene Flow Estimation on Large-scale Point Clouds, CVPR 2019, [Paper]
GA-Net: Guided Aggregation Net for End-to-end Stereo Matching, CVPR 2019, [Paper]
DPSNet: End-to-end Deep Plane Sweep Stereo, ICLR 2019, [Paper]
Fast Depth Densification for Occlusion-aware Augmented Reality, SIGGRAPH-Asia 2018, [Project Page], another version
Learning To Adapt For Stereo, CVPR 2019, [Paper]
Pyramid Stereo Matching Network, [Paper]
Bridging Stereo Matching and Optical Flow via Spatiotemporal Correspondence, [Paper]
Sparse Depth Completion, [Paper], RGB 图像辅助雷达深度估计
GASDA, CVPR 2019, [Paper]
MVSNet: Depth Inference for Unstructured Multi-view Stereo, [Paper]，非官方实现版本的 MVSNet
Stereo R-CNN based 3D Object Detection for Autonomous Driving, CVPR 2019, [Paper]
Real-time self-adaptive deep stereo, CVPR 2019, [Paper]
High Quality Monocular Depth Estimation via Transfer Learning, CVPR 2019, [Paper], [Project Page]
Group-wise Correlation Stereo Network, CVPR 2019, [Paper]
DeepMVS: Learning Multi-View Stereopsis, CVPR 2018, [Project Page]，多目深度估计
FlowNet 2.0: Evolution of Optical Flow Estimation with Deep Networks, CVPR 2017，深度学习光流恢复
StereoVision-ADCensus，深度恢复代码集合（ADCensus, SGBM, BM）
SegStereo: Exploiting Semantic Information for Disparity Estimation，探究语义信息在深度估计中的作用
Light Field Depth Estimation using GAN，利用 GAN 进行光场深度恢复
EV-FlowNet: Self-Supervised Optical Flow for Event-based Cameras, Proceedings of Robotics 2018, [Paper]
DF-Net: Unsupervised Joint Learning of Depth and Flow using Cross-Task Consistency, ECCV 2018, [Paper]
GeoNet: Unsupervised Learning of Dense Depth, Optical Flow and Camera Pose, CVPR 2018, [Paper]

3D & Graphic

PRNet: Self-Supervised Learning for Partial-to-Partial Registration, NeurIPS 2019
Learning to Reconstruct 3D Human Pose and Shape via Model-fitting in the Loop, ICCV 2019, [Paper], [Project Page]
Cross View Fusion for 3D Human Pose Estimation, ICCV 2019, [Paper]，跨视角 3D 位姿估计
MVF-Net: Multi-View 3D Face Morphable Model Regression，多视角 3D 人脸重建，[Paper]
KillingFusion
ReFusion: 3D Reconstruction in Dynamic Environments for RGB-D Cameras Exploiting Residuals, [Paper]
densebody_pytorch, [Paper]
Single-Image Piece-wise Planar 3D Reconstruction via Associative Embedding, CVPR 2019, [Paper]，单目 3D 重建
HorizonNet: Learning Room Layout with 1D Representation and Pano Stretch Data Augmentation, CVPR 2019, [Paper]，深度学习全景转 3D
Adaptive O-CNN: A Patch-based Deep Representation of 3D Shapes, SIGGRAPH Asia 2018, [Project Page]

Other Collections

chinese-independent-blogs，中文独立博客集锦
StructureFlow: Image Inpainting via Structure-aware Appearance Flow，图像 inpainting
free-books，互联网上的免费书籍
AcademicPages，通用的学术主页模版
MMdnn，实现深度学习模型之间的相互转换
tensorflow2caffemodel, TensorFlow 模型转 caffemodel
lihang-code,《统计学习方法》的代码实现
sse2neon, sse2neon, SSE 转 neon，嵌入式移植时可能会用到；
Production-Level-Deep-Learning，深度学习模型部署流程
动手学深度学习 Dive-into-DL-PyTorch
machine-learning-yearning-cn，Machine Learning Yearning 中文版 - 《机器学习训练秘籍》 - Andrew Ng 著
academicpages.github.io，学术主页模板
Coursera-ML-AndrewNg-Notes，吴恩达老师的机器学习课程个人笔记
machine-learning-notes，机器学习，概率模型和深度学习的讲义（1500+页）和视频链接
CNN-Visualization, CNN 可视化、理解 CNN
Awesome Semantic Segmentation，语义分割集合
IROS2018 SLAM Collections, IROS 2018 集合
VP-SLAM-SC-papers, Visual Positioning & SLAM & Spatial Cognition 论文统计与分析
Awesome System for Machine Learning
Machine-Learning-With-Python, 《机器学习实战》Python 代码实现
How to learn robotics，开源机器人学学习指南
Awesome Deep Vision, DL 在 CV 领域的应用
Single-Image-Super-Resolution，一个有关图像超分辨的合集
ai report, AI 相关的研究报告
State-of-the-art papers and code，搜集了目前 sota 的论文以及代码
CVPR 2019 (Papers/Codes/Project/Paper reading)
A curated list of papers & resources linked to 3D reconstruction from images，有关三维重建的论文汇总
SLAM-Jobs, SLAM/SFM 求职指南
Spatial Attentive Single-Image Deraining with a High Quality Real Rain Dataset, CVPR 2019，去雨
Densely Connected Pyramid Dehazing Network, CVPR 2018，去雾
MMSR，MMLAB 推出的超分辨工具箱
深度学习 OCR
西瓜书🍉学习笔记
awesome-reinforcement-learning-zh，强化学习从入门到放弃的资料
Deep Plug-and-Play Super-Resolution for Arbitrary Blur Kernels, CVPR 2019，超分辨
Cool Fashion Papers, Cool resources about Fashion + AI.
Deep Flow-Guided Video Inpainting, CVPR 2019, [Paper]，图像修复
YOLACT: Real-time Instance Segmentation
LPRNet: License Plate Recognition via Deep Neural Networks, [Paper]
CHINESE-OCR，运用 tf 实现自然场景文字检测
BeautyCamera，美颜相机，具有人脸检测、磨皮美白人脸、滤镜、调节图片、摄像功能
CV-arXiv-Daily，分享计算机视觉每天的 arXiv 文章
Pluralistic-Inpainting, ArXiv | Project Page | Online Demo | Video(demo)
An Interactive Introduction to Fourier Transforms，超棒的傅里叶变换图形化解释
pumpkin-book, 《机器学习》（西瓜书）公式推导解析
Julia
A Julia machine learning framework，一种基于 Julia 的机器学习框架
High-Performance Face Recognition Library on PyTorch，人脸识别库
Deep-Learning-Coursera，深度学习教程（deeplearning.ai）
The best resources around Machine Learning
VGGFace2: A dataset for recognising faces across pose and age
Statistical learning methods，统计学习方法
End-to-end Adversarial Learning for Generative Conversational Agents，2017，介绍了一种端到端的基于 GAN 的聊天机器人
Residual Non-local Attention Networks for Image Restoration, ICLR 2019.
MSGAN: Mode Seeking Generative Adversarial Networks for Diverse Image Synthesis, CVPR 2019, [Paper]
SPADE: Semantic Image Synthesis with Spatially-Adaptive Normalization, CVPR 2019, [Project Page]
Faceswap with Pytorch or DeepFake with Pytorch，换脸
DeepFaceLab，换脸

SLAM related

Pose/Object tracking

Depth/Disparity & Flow estimation

3D & Graphic

Other Collections

Comments