NVIDIA has opened source a comprehensive library of 3D deep learning based on pytorch

Click on the above “ Xiaobai studies vision ”, Optional plus " Star standard " or “ Roof placement ”

 Heavy dry goods , First time delivery 

edit ：Sophia

Reproduced in ：https://techxplore.com/news/2019-11-kaolin-comprehensive-library-d-deep.html

Because most real-world environments are three-dimensional , Therefore, deep learning models designed to analyze videos or complete tasks in real-world environments should also ideally be trained on 3D data . Such as robots 、 Autopilot 、 Technology tools like smartphones and other devices , At present, more and more 3D data are being generated , These data may eventually be processed by deep learning algorithms .

But so far , Only some AI researchers can use the necessary tools and platforms , This makes it difficult to train deep learning algorithms on massive 3D data ; In order to solve the problem of lack of ready-made tools ,NVIDIA A research team in has recently created a project called “ Kaolin ”（Kaolin） Of PyTorch Open source library , It aims to promote and promote the research of three-dimensional deep learning .

Address of thesis ：https://arxiv.org/pdf/1911.05063.pdf

A researcher who carried out this study said ：“ at present , There is no single open source software library that can support multiple forms of 3D data 、 Multiple tasks and evaluation criteria ; We decided to create Kaolin To fill this gap , This is the first comprehensive three-dimensional deep learning library .”

Kaolin Contains various tools for constructing deep learning architectures , Can analyze 3D data , And it has both effectiveness and ease of use , Allow researchers to use it to train deep learning algorithms , load 、 Preprocess and manipulate 3D data .Kaolin It also includes several graphics modules to edit 3D images （ Such as rendering 、 lighting 、 Shadows and view warps ）, And support a wide range of loss functions and evaluation indicators , So that researchers can easily evaluate their deep learning algorithm .

Whether for developers experienced in developing deep learning models , Or for developers who are just starting ,Kaolin Is a valuable tool . actually , Developers can also find several of the most advanced architectures in the Library , They can use it as a starting point or source of inspiration for their models . Although active 3D deep learning researchers will Kaolin As an effective means to speed up research , But newcomers to this field are also going to Kaolin As a starting point . In the future ,Kaolin It can also help developers accelerate 3D deep learning research , Assist developers in creating new AI architectures, training and evaluating them . meanwhile ,NVIDIA Of researchers are still planning to expand Kaolin And further enhance its functions , send Kaolin Become a one-stop platform for three-dimensional deep learning research .

The good news ！

Xiaobai learns visual knowledge about the planet

Open to the outside world

 download 1：OpenCV-Contrib Chinese version of extension module 

 stay 「 Xiaobai studies vision 」 Official account back office reply ： Extension module Chinese course , You can download the first copy of the whole network OpenCV Extension module tutorial Chinese version , Cover expansion module installation 、SFM Algorithm 、 Stereo vision 、 Target tracking 、 Biological vision 、 Super resolution processing and other more than 20 chapters .


 download 2：Python Visual combat project 52 speak 
 stay 「 Xiaobai studies vision 」 Official account back office reply ：Python Visual combat project , You can download, including image segmentation 、 Mask detection 、 Lane line detection 、 Vehicle count 、 Add Eyeliner 、 License plate recognition 、 Character recognition 、 Emotional tests 、 Text content extraction 、 Face recognition, etc 31 A visual combat project , Help fast school computer vision .


 download 3：OpenCV Actual project 20 speak 
 stay 「 Xiaobai studies vision 」 Official account back office reply ：OpenCV Actual project 20 speak , You can download the 20 Based on OpenCV Realization 20 A real project , Realization OpenCV Learn advanced .


 Communication group 

 Welcome to join the official account reader group to communicate with your colleagues , There are SLAM、 3 d visual 、 sensor 、 Autopilot 、 Computational photography 、 testing 、 Division 、 distinguish 、 Medical imaging 、GAN、 Wechat groups such as algorithm competition （ It will be subdivided gradually in the future ）, Please scan the following micro signal clustering , remarks ：” nickname + School / company + Research direction “, for example ：” Zhang San  +  Shanghai Jiaotong University  +  Vision SLAM“. Please note... According to the format , Otherwise, it will not pass . After successful addition, they will be invited to relevant wechat groups according to the research direction . Please do not send ads in the group , Or you'll be invited out , Thanks for your understanding ~