深度学习资料整理

主要收集了一些开源代码的链接，以便平时可以随时查看和学习。
(1)YOLOV3 tensorflow源码
https://github.com/YunYang1994/tensorflow-yolov3
基于Win10的TensorFlow环境配置

计算机平台：Win 10 x64
TensorFlow版本：1.15(cpu版)
Python版本：3.6.5
Pycharm版本：2017.3 professional

https://github.com/zzh8829/yolov3-tf2

(2)52 个深度学习目标检测模型汇总，论文、源码一应俱全
https://blog.csdn.net/weixin_42137700/article/details/104813112

(3)视频教程
https://ai.51cto.com/

(4)3D目标检测
汇总|3D目标检测文章(CVPR2020)：
https://blog.csdn.net/Yong_Qi2015/article/details/106913355

PCL点云障碍物检测：
https://github.com/williamhyin/SFND_Lidar_Obstacle_Detection

https://github.com/open-mmlab/OpenPCDet/

3D目标检测论文汇总(单目、激光、融合)
https://blog.csdn.net/qq_29462849/article/details/103543233
PointNet：
https://github.com/charlesq34/pointnet
https://github.com/fxia22/pointnet.pytorch
PointNet++：
https://github.com/charlesq34/pointnet2
https://github.com/yanx27/Pointnet_Pointnet2_pytorch
Lidar 3D目标检测
https://blog.csdn.net/weixin_40731977/article/details/89439198
https://github.com/sshaoshuai/PointRCNN
PointRCNN网络可视化，代码详解:
https://blog.csdn.net/wqwqqwqw1231/article/details/90788500
pointpillars:
用于训练：
https://github.com/SmallMunich/nutonomy_pointpillars.git
https://github.com/nutonomy/second.pytorch#pointpillars
https://github.com/qianguih/voxelnet
https://github.com/Hqss/VoxelNet_PyTorch
安装教程：https://blog.csdn.net/r1141207831/article/details/103000339
tensorrt部署：
https://githubmemory.com/repo/hyaihjq/PointPillars_MultiHead_40FPS
https://github.com/yukkysaito/autoware_perception
https://github.com/hova88/PointPillars_MultiHead_40FPS

CenterFusion：融合雷达与摄像头数据的高精度3D目标检测
https://segmentfault.com/a/1190000038424123
https://github.com/mrnabati/CenterFusion

(5)汇总 | 基于激光雷达的3D目标检测开源项目&数据集
https://www.sohu.com/a/421173574_100007727

(6)动手学无人驾驶（4）：基于激光雷达点云数据3D目标检测
PointRCNN
https://blog.csdn.net/cg129054036/article/details/105372783?utm_medium=distribute.pc_relevant.none-task-blog-baidujs_title-3&spm=1001.2101.3001.4242
【代码阅读】PointRCNN网络可视化，代码详解:
https://blog.csdn.net/wqwqqwqw1231/article/details/90788500

(7)NASA的火星车开源项目
https://github.com/nasa-jpl/open-source-rover

(8)百度大脑
https://aistudio.baidu.com/aistudio/index

(9)tensorflow源码
https://github.com/tensorflow/tensorflow.git
TensorFlow Lite:
https://tensorflow.google.cn/lite?hl=zh-cn
在ARM板子上测试Tensorflow Lite
https://www.ctolib.com/topics-132100.html

pytorch源码
https://github.com/pytorch/pytorch

PyTorch官方教程中文版：
https://pytorch123.com/

(10)跟踪算法总结
https://blog.csdn.net/wb790238030/article/details/89326144
https://github.com/bilylee/siamfc-tensorflow

多目标跟踪：
https://zhuanlan.zhihu.com/visual-tracking

动手学无人驾驶（3）：基于激光雷达3D多目标追踪：
https://blog.csdn.net/cg129054036/article/details/103567265

Top1的3D目标检测方法（已开源）
https://yongqi.blog.csdn.net/article/details/107804981

(11)CVPR 2020 论文大盘点-人体姿态估计与动作捕捉篇
https://zhuanlan.zhihu.com/p/149340496

(12)行人检测算法(深度学习、机器学习)
https://zhuanlan.zhihu.com/p/37468092
https://zhuanlan.zhihu.com/p/267875460

行人重识别算法SCPNet论文解读与代码开源：
https://zhuanlan.zhihu.com/p/56418084

(13)代博博客
https://www.cnblogs.com/dzyBK/

(14)MSCKF
https://zhuanlan.zhihu.com/p/304889273
https://www.zhihu.com/people/mao-shu-yuan

(15)vins-mono
https://zhuanlan.zhihu.com/p/150716308

(16)语义分割
https://www.cnblogs.com/xiexiaokui/p/12152165.html
https://github.com/matterport/Mask_RCNN
教程：https://xiaosongshine.blog.csdn.net/article/details/99670519

道路分割：
https://github.com/MaybeShewill-CV/lanenet-lane-detection
https://github.com/klintan/pytorch-lanenet

可行驶区域分割
https://github.com/hlwang1124/SNE-RoadSeg

(17)人体姿态估计
https://github.com/cbsudux/awesome-human-pose-estimation
https://blog.csdn.net/m0_37644085/article/details/82021848

(18)超快目标检测
1.3MB超轻YOLO：
https://github.com/dog-qiuqiu/Yolo-Fastest

比SSD效果更好的MobileNet-YOLO：
https://github.com/dog-qiuqiu/MobileNet-Yolo

NanoDet（超轻量，速度很快）
https://guo-pu.blog.csdn.net/article/details/110410940
https://github.com/RangiLyu/nanodet
训练自己数据集：
https://blog.csdn.net/qq_39056987/article/details/112983966
VOC格式标注转COCO格式：
https://blog.csdn.net/qq_39056987/article/details/112358124

Yolov4-tiny
https://blog.csdn.net/jiejinquanil/article/details/106998409
yolov4-tiny训练自己的模型并部署到RK的NPU上推理：
https://blog.csdn.net/zengwubbb/article/details/109134859?utm_medium=distribute.pc_relevant.none-task-blog-BlogCommendFromMachineLearnPai2-2.control&depth_1-utm_source=distribute.pc_relevant.none-task-blog-BlogCommendFromMachineLearnPai2-2.control
训练：
https://blog.csdn.net/qq_32335689/article/details/109995489
https://blog.csdn.net/u010881576/article/details/107053328/
https://github.com/bubbliiiing/yolov4-tiny-pytorch
YOLOV4与YOLOV3的区别：
https://blog.csdn.net/weixin_42206075/article/details/113352277

官方链接
yoloV4论文：《YOLOv4: Optimal Speed and Accuracy of Object Detection》
yoloV4 Github代码：https://github.com/AlexeyAB/darknet
yoloV1~V3 Github代码:https://github.com/pjreddie/darknet
darknet官网：https://pjreddie.com/darknet/

yolov5原作者链接：
https://github.com/ultralytics/yolov5
yolov5系列模型转换：
https://github.com/wang-xinyu/tensorrtx
YOLOV5测试及训练自己的数据集：
https://blog.csdn.net/weixin_43871135/article/details/106803636

【玩转yolov5】使用bdd100k数据集训练行人和全车模型：
https://blog.csdn.net/ChuiGeDaQiQiu/article/details/113415081?utm_medium=distribute.pc_aggpage_search_result.none-task-blog-2_{aggregatepage}first_rank_v2~rank_aggregation-6-113415081.pc_agg_rank_aggregation&utm_term=bdd100k+%E8%AE%AD%E7%BB%83&spm=1000.2123.3001.4430

(19)数据集
https://blog.csdn.net/qq_30460949/article/details/107593347

BDD100K：一个大规模、多样化的驾驶视频数据集：
https://blog.csdn.net/jocelyn870/article/details/81207448

(20)Windows下Git的使用
https://www.cnblogs.com/gdjlc/archive/2019/12/23/12088872.html

(21)激光雷达建图
loam：
https://github.com/laboshinl/loam_velodyne
gmapping：
从应用gmapping包来讲，ROS中的slam_gmapping包也是调用了openslam_gmapping开源算法，gmapping的代码分为两部分：
https://github.com/ros-perception/openslam_gmapping
https://github.com/ros-perception/slam_gmapping
GMapping原理分析：
https://blog.csdn.net/liuyanpeng12333/article/details/81946841
gmapping 公式推导：
https://blog.csdn.net/Snall_Qiu/article/details/108088680

(22)CornerNet-Lite
https://www.bilibili.com/video/av540024019/
https://github.com/princeton-vl/CornerNet-Lite

(23)轨迹预测
http://apolloscape.auto/trajectory.html

Social-lstm算是比较出名的行人轨迹预测方法
https://zhuanlan.zhihu.com/p/63864194

https://github.com/HaoyunHong/Spectral-Trajectory-and-Behavior-Prediction

https://zhuanlan.zhihu.com/p/139905513

(24)Apollo代码
https://github.com/ApolloAuto/apollo
代码分析：
https://gitee.com/wan9hao/Dig-into-Apollo
阿波罗
http://apollo.auto

(25)车道线检测
https://github.com/cfzd/Ultra-Fast-Lane-Detection

(26)语义分割BiSeNet
https://github.com/CoinCheung/BiSeNet

(27)图像分类网络
https://github.com/lxztju/pytorch_classification

(28)autoware
基于ros1：
https://www.autoware.ai/
https://github.com/Autoware-AI
https://github.com/tier4/AutowareArchitectureProposal.proj
基于ros2：
https://www.autoware.auto/
https://gitlab.com/autowarefoundation/autoware.auto/AutowareAuto

(29)TensorRT
https://github.com/NVIDIA/TensorRT/
https://github.com/NVIDIA-AI-IOT
https://github.com/NVIDIA-AI-IOT/CUDA-PointPillars

(30)onnx
https://github.com/onnx
https://github.com/onnx/onnx-tensorrt

(31)LibTorch
https://pytorch.org/
https://github.com/yasenh/libtorch-yolov5
libtorch版本：libtorch-cxx11-abi-shared-with-deps-1.9.1+cu102
libtorch 常用api函数示例（史上最全、最详细）：
https://www.cnblogs.com/yanghailin/p/12901586.html

(31)YOLOX
https://zhuanlan.zhihu.com/p/392570215

(32)3D激光建图
https://github.com/koide3/hdl_graph_slam