当前位置:网站首页>Pointnext: review pointnet through improved model training and scaling strategies++
Pointnext: review pointnet through improved model training and scaling strategies++
2022-07-04 19:56:00 【Master Ma】
PointNet++ It is one of the most influential neural architectures for point cloud understanding . Even though PointNet++ The accuracy of has been PointMLP and Point Transformer Wait until the recent network has largely surpassed , But we found that a large part of the performance improvement is due to the improvement of training strategies , That is, data enhancement and optimization technology , And increased model size rather than architectural innovation . therefore ,PointNet++ Its full potential remains to be explored .
In this work , We reexamine the classical model by systematically studying the model training and scaling strategies PointNet++, And provides two main contributions .
First , We propose a set of improved training strategies , Significantly improved PointNet++ Performance of . for example , We show that , Without changing the architecture ,PointNet++ stay ScanObjectNN The overall accuracy of object classification (OA) It can be downloaded from 77.9% Up to 86.1%, Even better than the most advanced PointMLP.
secondly , We will invert residual bottleneck design and separability MLP introduce PointNet++, To achieve efficient and effective model scaling , And to put forward PointNeXt, That is, the next version of PointNets.
PointNeXt It can be flexibly expanded , stay 3D Both classification and segmentation tasks are superior to the most advanced methods .
Thesis title :PointNeXt: Revisiting PointNet++ with Improved Training and Scaling Strategies
Detailed interpretation :
https://www.aminer.cn/research_report/62afdfb27cb68b460fd6c4fa?download=false
icon-default.png?t=M5H6https://www.aminer.cn/research_report/62afdfb27cb68b460fd6c4fa?download=false
AMiner link :https://www.aminer.cn/?f=cs
https://zhuanlan.zhihu.com/p/526818590
This paper re examines the classic PointNet++, And provides two main contributions , And then put forward PointNeXt, performance SOTA! Better performance than the PointMLP、Point Transformer Wait for the Internet , The code is open source (5 God 90+ star)!
PointNeXt: Revisiting PointNet++ with Improved Training and Scaling Strategies
Company :KAUST, Microsoft
Code :https://github.com/guochengqian/pointnext
The paper :https://www.aminer.cn/pub/62a2b6955aee126c0f4d8e79
PointNet++ It is one of the most influential neural architectures for point cloud understanding . Even though PointNet++ The accuracy of has been PointMLP and Point Transformer Wait until the recent network has largely surpassed , But we found that a large part of the performance improvement is due to the improvement of training strategies , That is, data enhancement and optimization technology , And increased model size rather than architectural innovation . therefore ,PointNet++ Its full potential remains to be explored .
In this work , We reexamine the classical model by systematically studying the model training and scaling strategies PointNet++, And provides two main contributions .
First , We propose a group Improved training strategies , Significantly improved PointNet++ Performance of . for example , We show that , Without changing the architecture ,PointNet++ stay ScanObjectNN The overall accuracy of object classification (OA) It can be downloaded from 77.9% Up to 86.1%, Even better than the most advanced PointMLP.
secondly , We will Inverted residual bottleneck design and separability MLP introduce PointNet++, To achieve efficient and effective model scaling , And to put forward PointNeXt, That is, the next version of PointNets.
PointNeXt It can be flexibly expanded , stay 3D Both classification and segmentation tasks are superior to the most advanced methods .
In this section , We showed how to improve through more advanced training strategies and model scaling strategies PointNet++ Performance of . We will introduce them in two sections :
(1) Modernization of training strategies ;
(2) Modernization of network architecture .
Modernization of training strategies
In this chapter , We briefly describe our research methods , Specific training strategies can be seen in the subsequent Ablation Experiment chapter .
Data to enhance
Data enhancement is one of the most important methods to improve the performance of neural networks , and PointNet++ Simple data enhancement combinations such as random rotation are used , The zoom , translation , shake (jitter) And applied to different data sets . Some of the latest methods use stronger data enhancement methods . for example , KPConv Random deactivation during training (drop) Some color information . In this work , We collected common data enhancement methods used in recent methods , The effect of each data enhancement method on each data set is quantitatively studied by superposition experiment . For each data set , We propose a set of improved data enhancement methods , It can be greatly improved PointNet++ Performance of .
Optimization strategy
Optimization techniques mainly include loss functions (loss function), Optimizer (optimizer), Learning rate Planner (learning rate schedulers), And super parameters (hyperparmeters). With the development of machine learning theory , Modern neural networks can be theoretically better ** Optimizer ( Such as AdamW) And a better loss function (CrossEntropy with label smoothing)** Training .Cosine learning rate decay It has also been widely used in recent years , Because compare step decay, It is simpler to adjust parameters and the effect is not bad . In this work , We quantify the effects of each optimization strategy on PointNet++ Influence . alike , For each data set , We propose a set of improved optimization techniques that can further improve network performance .
Modernization of model architecture : Minor modifications → Great improvement
Receptive field scaling
In point cloud network , Use different ball query radius ( Query radius ) Will affect the receptive field of the model , Which in turn affects performance . We find that the initial radius has a great influence on the network performance , And the optimal query radius is different on different datasets . Besides , We find relative coordinates Make network optimization more difficult , Performance degradation . therefore , We propose to use the relative coordinates to query the radius to achieve the normalization :
If there is no normalization , The value of relative coordinates will be very small ( Less than the radius ). This requires that the network can learn to apply greater weight to . This makes optimization difficult , In particular, the regularization method considering the weight attenuation limits the size of the network weight .
Model scaling
PointNet++ The scale of models used for classification and segmentation is smaller than 2M. Now the network parameters are generally 10M above [3,4]. Interestingly , We found that no matter using more SA The module is still larger channel size Will not significantly improve accuracy , But it leads to thoughput A marked decline . This is mainly caused by gradient vanishing and over fitting . In this section , We proposed Inverted Residual MLP (InvResMLP) Module to achieve efficient and practical model scaling . The module is based on SA On module , As shown in the middle of Figure 1 .InvResMLP and SA There are three differences between modules :
- A residual connection is added between the input and output of the module , To alleviate the gradient disappearance problem
- Separable MLP To reduce the amount of calculation , And enhance point by point feature extraction
- introduce invertedbottleneck The design of the , To improve the ability of feature extraction
stay PointNet++ On the basis of InvResMLP And the macro structure changes shown in Figure 1 , We proposed PointNeXt. We will stem MLP Of channel The size is expressed as C, take InvResMLP The number of modules is expressed as B. We PointNeXt The configuration of the series is summarized as follows :
- PointNeXt-S: C = 32, B = 0
- PointNeXt-B: C = 32, B = (1, 2, 1, 1)
PointNeXt-L: C = 32, B = (2, 4, 2, 2)
PointNeXt-XL: C = 64, B = (3, 6, 3, 3)
experiment
stay S3DIS Semantic segmentation ,PointNeXt-XL With mIoU/OA/mACC=74.9%/90.3%/83.0% Beyond Point Transformer obtain SOTA Performance and faster reasoning . stay ScanObjectNN Classification ,PointNeXt-S Beyond the current SOTA Method PointMLP, And the reasoning speed is ten times faster . stay ShapeNetPart Partial segmentation , Widened model PointNeXt-S(C=160) achieve 87.2 Instance mIoU, transcend SOTA CurNet.
Ablation Experiment
边栏推荐
- Comment utiliser async awati asynchrone Task Handling au lieu de backgroundworker?
- HMM hidden Markov model and code implementation
- Multi table operation - external connection query
- Kotlin basic data type
- 公司要上监控,Zabbix 和 Prometheus 怎么选?这么选准没错!
- kotlin 循环控制
- 偏移量函数及开窗函数
- How test engineers "attack the city" (Part I)
- 多表操作-外连接查询
- 1007 maximum subsequence sum (25 points) (PAT class a)
猜你喜欢
西门子HMI下载时提示缺少面板映像解决方案
PointNeXt:通过改进的模型训练和缩放策略审视PointNet++
Dark horse programmer - software testing - 09 stage 2-linux and database -31-43 instructions issued by modifying the file permission letter, - find the link to modify the file, find the file command,
Upgrade the smart switch, how much is the difference between the "zero fire version" and "single fire" wiring methods?
线上数据库迁移的几种方法
黑马程序员-软件测试--09阶段2-linux和数据库-31-43修改文件权限字母发的说明,-查找链接修改文件,查找文件命令,链接文件,压缩解压方式,vi编辑器基本使用,
Lm10 cosine wave homeopathic grid strategy
FPGA timing constraint sharing 01_ Brief description of the four steps
C# 使用StopWatch测量程序运行时间
上线首月,这家露营地游客好评率高达99.9%!他是怎么做到的?
随机推荐
kotlin 基本数据类型
C语言-入门-基础-语法-流程控制(七)
Find the nth power of 2
Pythagorean number law (any three numbers can meet the conditions of Pythagorean theorem)
1002. A+b for Polynomials (25) (PAT class a)
Is it safe to open an account at Great Wall Securities? How to open an account when buying stocks
Educational Codeforces Round 22 E. Army Creation
The explain statement in MySQL queries whether SQL is indexed, and several types in extra collate and summarize
Swagger突然发癫
Pytorch学习(四)
HDU 1097 A hard puzzle
Detailed explanation of the binary processing function threshold() of opencv
Euler function
Abc229 summary (connected component count of the longest continuous character graph in the interval)
Mysql database basic operation -ddl | dark horse programmer
1003 Emergency(25 分)(PAT甲级)
1008 elevator (20 points) (PAT class a)
如何使用Async-Awati异步任务处理代替BackgroundWorker?
HMM hidden Markov model and code implementation
SSRS筛选器的IN运算(即包含于)用法