yannzyl / apollo-note Goto Github PK
View Code? Open in Web Editor NEWNote for Apollo 3.0 perception, prediction and planning modules
Home Page: https://github.com/YannZyl/Apollo-Note
Note for Apollo 3.0 perception, prediction and planning modules
Home Page: https://github.com/YannZyl/Apollo-Note
@YannZyl 您好!非常感谢您的分享,学到很多。有一个问题问一下,在obstacles_lidar_2_cnn.md中分析的并查集部分中的Traverse函数,“每棵树只有一条path的is_center=true", 其他都是false,导致后续合并时一棵树只与另一棵树的指定支路去合并。” 不太明白为什么不把所有的path的is_center设为true,这样相近的所有树及分支都能够合并,只与一个path合并的话,不会漏掉与这一个path距离远而与其他path距离较近的同一个物体上的点吗?如果我想尝试合并所有相近的树和分支,是不是应该把if (x->traversed == 2)改为if(x->traversed == 2 || x->traversed == 1)?
谢谢!
Thank you very much for your summary and share.I would like to ask two questions in the Apollo/ Percetion/LIdar module:
The following source code, the path for "modules/perception/obstacle/lidar/tracker/hm_tracker/track_object_distance. cc" , when lidar track ComputeAssociateMatrixhas five factors, including: location_distance,direction_distance,bbox_size_distance,point_num_distance,histogram_distance. I want to ask whether there is any selection basis for selecting the 5 parameters for weighting. There are 5 default weights s_location_distance_weight_, s_direction_distance_weight_, s_bbox_size_distance_weight_, s_point_num_distance_weight_, s_histogram_distance_weight_ are where the selection is given**?
float TrackObjectDistance::ComputeDistance( ObjectTrackPtr track, const Eigen::VectorXf& track_predict, const std::shared_ptr<TrackedObject>& new_object) { // Compute distance for given trakc & object ....... float result_distance = s_location_distance_weight_ * location_distance + s_direction_distance_weight_ * direction_distance + s_bbox_size_distance_weight_ * bbox_size_distance + s_point_num_distance_weight_ * point_num_distance + s_histogram_distance_weight_ * histogram_distance; return result_distance; }
In the source code, association_(i, j) <= connected_threshold==4. What factors are considered in the selection of this connected_threshold=4?
Looking forward to your reply and best wishes!
您好,看到在CNN网络的输出中,由两项数据是表示中心偏移,想请问,如果自己标注数据训练,这个中心偏移该如何标注呢?
Which model is used for traffic light perception ?
Hi, may I ask how to convert the minbox direction and center into a Pose (with quarterion and dimension)?
Simply I know the l,w,h, means, but what does direction means? Coordinates terms to lidar origin? How to get the pose of obstacle in quarterion?
数据关联部分,在ConnectedComponentAnalysis中,得到的components二维向量中,每一行为一个子图的组成元素,这个子图是如何划分的,有点不明白
A declarative, efficient, and flexible JavaScript library for building user interfaces.
🖖 Vue.js is a progressive, incrementally-adoptable JavaScript framework for building UI on the web.
TypeScript is a superset of JavaScript that compiles to clean JavaScript output.
An Open Source Machine Learning Framework for Everyone
The Web framework for perfectionists with deadlines.
A PHP framework for web artisans
Bring data to life with SVG, Canvas and HTML. 📊📈🎉
JavaScript (JS) is a lightweight interpreted programming language with first-class functions.
Some thing interesting about web. New door for the world.
A server is a program made to process requests and deliver data to clients.
Machine learning is a way of modeling and interpreting data that allows a piece of software to respond intelligently.
Some thing interesting about visualization, use data art
Some thing interesting about game, make everyone happy.
We are working to build community through open source technology. NB: members must have two-factor auth.
Open source projects and samples from Microsoft.
Google ❤️ Open Source for everyone.
Alibaba Open Source for everyone
Data-Driven Documents codes.
China tencent open source team.