3D目标检测应该怎么开始学习?
究极路线:
https://github.com/TianhaoFu/Awesome-3D-Object-Detection/
食用方法:
1. 大致扫完一眼后,快速进入blog部分,学习经典方法。
2. 进入course部分,学习图宾根大学课程的3d检测部分。
3. 进入video部分,看3d检测相关的seminar。
4. 现在,你已经大致具备3d检测领域的intuition了,之后
5. 如果你想发论文,你可以进入paper部分。
6. 如果你想打比赛,你可以进入competition solution部分。
7. 如果你想做工程,你可以进入engineering部分。
3D视觉工坊课程链接
3D缺陷检测课程
学习方法:
看论文和对应代码,mmdetection3d和openpcdet, 可以先从这些codebase入手.
经典算法,比如SECOND,PillarNet,VoxelNet。
VoxelNet阅读笔记
mmdetection3D入门介绍
- 首先推荐学习这两篇mmdetection3d教程:
- 其次要了解mmdet3d的主要依赖package mmdetection(以下简称mmdet):
- 最后要熟悉runner,hook等结构,参考mmcv教程:
【综述】PointNet、PointNet++、 F-PointNet基于深度学习的3D点云分类和分割
-
PointNet: Deep Learning on Point Sets for 3D Classificationand Segmentation
-
PointNet++: Deep Hierarchical Feature Learning on Point Setsin a Metric Space
-
F-PointNet:FrustumPointNets for 3D Object Detection from RGB-D Data
双愚:
PCL:
- 主要论文:
- PointRCNN:3D Object Proposal Generation and Detection from Point Cloud
- PointPillar:Fast Encoders for Object Detection from Point Clouds
- PointPainting:Sequential Fusion for 3D Object Detection
- 相关论文:
- PointNet:Deep Learning on Point Sets for 3D Classification and Segmentation
- PointNet++:Deep Hierarchical Feature Learning on Point Sets in a Metric Space
- YOLO3D: End-to-end real-time 3D Oriented Object Bounding Box Detection from LiDARPoint Cloud
- Complex-YOLO: An Euler-Region-Proposal for Real-time 3D Object Detection on Point Clouds
- VoxelNet: End-to-End Learning for Point Cloud Based 3D Object Detection
- SECOND: Sparsely Embedded Convolutional Detection
- PV-RCNN: Point-Voxel Feature Set Abstraction for 3D Object Detection
- 3DSSD: Point-based 3D Single Stage Object Detector
- Structure Aware Single-stage 3D Object Detection from Point Cloud
- CIA-SSD: Confifident IoU-Aware Single-Stage Object Detector From Point Cloud
- Joint 3D Proposal Generation and Object Detection from View Aggregation
- Multi-View 3D Object Detection Network for Autonomous Driving
- EPNet: Enhancing Point Features with Image Semantics for 3D Object Detection
- CLOCs: Camera-LiDAR Object Candidates Fusion for 3D Object Detection
- PanoNet3D: Combining Semantic and Geometric Understanding for LiDAR Point Cloud Detection
- Efficient-3DCNNs:Resource Efficient 3D Convolutional Neural Networks
- 星球学习资料1
- 星球学习资料2
- 星球学习资料3
- 星球学习资料4-3D目标检测
- 主要针对3D object相关算法进行了汇总
- 主要针对3D object相关论文进行了汇总
- 3D-Machine-Learning
- 星球-3D目标检测代码大全
- 星球-BEV感知
- 星球-轨迹预测
- 星球-全栈路线
- Deploying CenterNet model using NVIDIA DeepStream
- TRTForYolov3
- PointPillars Inference with TensorRT
- 在这个project中,提供大量的cuda和tensorrt学习案例
- cmake-examples入门教程
- 介绍如何生成可执行文件,库文件,添加依赖头文件,库文件,包,安装,测试
- Hi3516
- 海思Hi3516DV300移植YOLOv3
NLP: