Skip to content

An efficient implementation of KinectFusion with modern C++ and CUDA.

License

Notifications You must be signed in to change notification settings

isikmustafa/kinectfusion-cuda

Folders and files

NameName
Last commit message
Last commit date

Latest commit

 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 

Repository files navigation

An implementation of KinectFusion on CUDA platform.

How to build?

  • Clone the repository recursively (to also clone the submodules)
  • Go here https://www.libsdl.org/download-2.0.php to download precompiled dynamic and static libraries. Put them under 3rd\SDL2-2.0.9\libs. You can build the SDL if the precompiled libraries are not suitable for you.
  • Open the visual studio project (works fine with VS 2017). Build and run it.
  • We were able process a frame around 25 milliseconds on our Nvidia GTX 960M graphics card.

Some pictures of reconstructed kitchen:

alt text

alt text

alt text

About

An efficient implementation of KinectFusion with modern C++ and CUDA.

Resources

License

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published