Forums - Computer Vision for Windows on ARM (WinRT)

2 posts / 0 new
or Register
Last post
Computer Vision for Windows on ARM (WinRT)
Annie
Join Date: 5 Nov 12
Posts: 2
Posted: Mon, 2012-11-05 02:48
Top

   (redirected from http://stackoverflow.com/q/13230147/1712065)

   I am learning FastCV (Alpha) for Windows RT. According to Getting Started guide:

The first implementation is designed to run efficiently on any ARM processor, and is named ‘FastCV for ARM.’ The second FastCV implementation runs only on Qualcomm SoC’s and is called ‘FastCV for Snapdragon.’ ‘FastCV for Snapdragon’ has the same API as ‘FastCV for ARM,’ but it provides hardware accelerated implementations of the CV functions included in the library.

   Then the metro lib integration guide say:

Configuration Manager as highlighted should be set to Release Mode (metro lib is compiled for this mode) and the platform should be set to ARM. If ARM does not appear by default, click add new platform -> ARM in configuration manager..

   Questions:

  • Does it mean we cannot consume this library on Windows Store apps running on ia/x86 systems?
  • Would FastCV code run only on Snapdragon or other ARM processors too (like Tegra 3)?
  • Is there any other implementation of CV for OCR on Windows RT?
  • Up0
  • Down0
Re: Computer Vision for Windows on ARM (WinRT)
jeff4s Moderator
Join Date: 4 Nov 12
Posts: 106
Posted: Tue, 2012-11-06 14:12
Top
  • Does it mean we cannot consume this library on Windows Store apps running on ia/x86 systems?

You are right. The library cannot be used on x86 systems as it's compiled specifically for Windows on ARM platforms.

  • Would FastCV code run only on Snapdragon or other ARM processors too (like Tegra 3)?

FastCV for WinRT would run on all ARM processors but Snapdragon processors would give much better performance.

  • Is there any other implementation of CV for OCR on Windows RT?

FastCV does not provide a list of APIs specifically for OCR. Instead, FastCV  provides generic low-level functions that developers can identify and use to accelerate their specific OCR implementations.

 

  • Up0
  • Down0
or Register

Opinions expressed in the content posted here are the personal opinions of the original authors, and do not necessarily reflect those of Qualcomm Incorporated or its subsidiaries (“Qualcomm”). The content is provided for informational purposes only and is not meant to be an endorsement or representation by Qualcomm or any other party. This site may also provide links or references to non-Qualcomm sites and resources. Qualcomm makes no representations, warranties, or other commitments whatsoever about any non-Qualcomm sites or third-party resources that may be referenced, accessible from, or linked to this site.