Skip to content

Using CityScapes to perform semantic segmentation in the AV space. Optimizing for FPS by leveraging PyTorch's C++ API (LibTorch) and NVIDIA's CV-CUDA package to run these models at the highest FPS possible!

Notifications You must be signed in to change notification settings

srianumakonda/Semantic-Segmentation-CUDA

Folders and files

NameName
Last commit message
Last commit date

Latest commit

 

History

29 Commits
 
 
 
 
 
 
 
 
 
 
 
 
 
 

Repository files navigation

To get started with this project, simply git clone the repository and ________.

When building from source with OpenCV in an anaconda venv, make sure that you add the CUDA and CUDNN adresses to PATH. This is absolutely essential to ensure that you can make the most out of your GPU and ensure that you can run at the highest FPS possible.

To do this, open bashrc by typing ~/.bashrc in terminal and add the PATH to the end of the file.

Demo gif of training over time:

alt text

About

Using CityScapes to perform semantic segmentation in the AV space. Optimizing for FPS by leveraging PyTorch's C++ API (LibTorch) and NVIDIA's CV-CUDA package to run these models at the highest FPS possible!

Resources

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published

Languages