xiaojake Goto Github PK

minisam_lib

Lightweighted graph optimization (Factor graph) library.

minklocmultimodal

!!激光雷达和单目图像融合的位置识别方法，性能SOTA__MinkLoc++: Lidar and Monocular Image Fusion for Place Recognition

mixformer

!!基于transform的端到端目标跟踪新网络 MixFormer: End-to-End Tracking with Iterative Mixed Attention

mlcc

Fast and Accurate Extrinsic Calibration for Multiple LiDARs and Cameras

mmdetection3d

!3D目标检测算法OpenMMLab's next-generation platform for general 3D object detection.

mockm

!集模拟和调试功能于一身. 自动生成数据和增删查改一系列接口, 支持文件上传下载, 延迟, 自定义接口逻辑...

monodtr

!基于Transformer的单目3D目标检测网络MonoDTR: Monocular 3D Object Detection with Depth-Aware Transformer (CVPR 2022)

monorec

!MonoRec：无需激光雷达，单相机实现三维场景的稠密重建，去除运动伪影，泛化性能好 Official implementation of the paper: MonoRec: Semi-Supervised Dense Reconstruction in Dynamic Environments from a Single Moving Camera (CVPR 2021)

mpn

!使用元原型网络学习视频中的正常动态:Official codes of CVPR21 paper: Learning Normal Dynamics in Videos with Meta Prototype Network

mrs_uav_system

!无人机多机控制,估计,仿真系统:The entry point to the MRS UAV system

msc-vo

MSC-VO: 基于曼哈顿和结构约束的视觉里程计-This repository includes an RGB-D VO system that combines both point and line features and leverages, if exist, structural regularities and the Manhattan axes of the scene.

msckf_vio

!基于双目相机的msckf_vio: Robust Stereo Visual Inertial Odometry for Fast Autonomous Flight

mspfn

!图像去噪-图像去除雨雾影响，用于单幅图像去噪的多尺度渐进融合网络Multi-Scale Progressive Fusion Network for Single Image Deraining

mttr

!多模态Transformer用于视频分割,包括重叠对象,MTTR

mulls

!!多度量线性最小二乘法的多功能激光雷达SLAM,单激光雷达MULLS: Versatile LiDAR SLAM via Multi-metric Linear Least Square [ICRA '21]

multi-task-refinenet

!使用不对称的注释数据集，实时联合进行语义分割和深度估计，处理速度超快:Multi-Task (Joint Segmentation / Depth / Surface Normas) Real-Time Light-Weight RefineNet

multi_lidar_calib

!待开源,针对多个小fov固态激光雷达 不依赖参照物的外参标定

multi_lidar_calibration

!!多激光雷达外参自动标定:Multi lidar calibration tool from autoware, using NDT algorithm, need a approximate initial transformation

multiple-cameras-and-3d-lidars-extrinsic-calibration

基于全景基础设施仅使用单帧数据完成多相机/激光雷达间外参校准

mvision

机器人视觉 无人驾驶 视觉SLAM ORB LSD SVO DSO 深度学习目标检测yolov3 行为检测 opencv PCL 双目视觉

mvp

一种将RGB传感器无缝融合到基于激光雷达的3D目标检测之中的框架

my_solidworks

myworksshow

我的工作成果细节展示

ndt_localizer

!!!激光雷达为主的多传感器融合定位:NDT/pc_map/w_odom/imu

ndtmc-lio-sam

其中的回环效果可适应地下停车场

neor_mini

ROS-based Ackerman-like unmanned car.

neural-rgbd-surface-reconstruction

RGB-D表面重建:Official implementation of the CVPR 2022 Paper "Neural RGB-D Surface Reconstruction"

neuralrecon

!单目实时三维重建新框架NeuralRecon: Code for "NeuralRecon: Real-Time Coherent 3D Reconstruction from Monocular Video", CVPR 2021 oral

nice-slam

一种隐式神经可扩展编码RGB-D SLAM-[CVPR'22] NICE-SLAM: Neural Implicit Scalable Encoding for SLAM

oa-slam

结合物体语义进行相机重定位的vslam