Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Jump to bottom

UMat: support processing of ND arrays #21137

Open
alalek opened this issue Nov 27, 2021 · 0 comments
Open

UMat: support processing of ND arrays #21137

alalek opened this issue Nov 27, 2021 · 0 comments
Labels
category: core category: ocl category: t-api cv::UMat related issues feature optimization

Comments

@alalek
Copy link
Member

alalek commented Nov 27, 2021 •
edited

Epic ticket to track support of multidimensional arrays (ND-tensors) processing

They are widely used e.g., in DNN module.

We have gaps in these areas:

  • element-wise "arithm_op"
  • element-wise convertTo()
  • element-wise convertToFp16()
  • reduce functions (min/max/etc)
  • norm
  • ...

Current approach is fallback on CPU code (which cause performance issues).

There are two levels of implementations complexity:

  • support contiguous ND arrays
  • support ROI in ND arrays
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Assignees
No one assigned
Labels
category: core category: ocl category: t-api cv::UMat related issues feature optimization
Projects
None yet
Milestone
No milestone
Development

No branches or pull requests
1 participant
@alalek