Skip to content

enlenama/VideoFrame2Tensor

Folders and files

NameName
Last commit message
Last commit date

Latest commit

 

History

7 Commits
 
 
 
 
 
 

Repository files navigation

VideoFrame2Tensor

Overview

VideoFrame2Tensor provides an efficient pipeline for importing video frames directly into machine learning tensors, eliminating CPU-GPU data copies and big buffer transfers. It leverages hardware-accelerated preprocessing using D3D12 APIs and DirectML operators for real-time ML applications on Windows.

Directory Structure

  • explainer.md — Technical explainer and API usage details.
  • example/semantic_segmentation/ — Example implementation for real-time video semantic segmentation, it's a subset of WebNN Samples

About

No description, website, or topics provided.

Resources

Stars

0 stars

Watchers

1 watching

Forks

Releases

No releases published

Packages

 
 
 

Contributors