Because of the holidays, TI E2E™ design support forum responses will be delayed from Dec. 25 through Jan. 2. Thank you for your patience.

  • State Resolved
  • Locked Locked
  • Replies 2 replies
  • Subscribers 101 subscribers
  • Views 183 views
  • Users 0 members are here
Support feedback
Options
Options
Similar topics

This thread has been locked.

If you have a related question, please click the "Ask a related question" button in the top right corner. The newly created question will be automatically linked to this question.

TDA4VM: Use of GPU

KONG XIANGXU
Expert 1501 points
Part Number: TDA4VM

Hi,

    In the previous development, we used OpenCV to do panoramic image Mosaic.

    So, I wonder: would the use of the sdk have made a big difference if I had switched panorama concatenation to OpenGL? Is there any relevant case available?

    At the same time, if you change the OpenGL sdk to do panoramic stitching, how can you intuitively know that the GPU has been used for acceleration in the process?

    Regards，

    Kong

  • +1
    TI__Mastermind 36307 points

    Hello Kong,

    Yes, OpenCV "concatenation" of the images would probably use the CPU for the stitching. You could use OpenGL to do the same, however, we do not have a reference algorithm at the moment. If you can find the algorithm implemented in any of the APIs supported by our SDK for GPU, you could run it there.

    If you did change the algorithm for panoramic stitching to use the GPU, intuitively you could check the GPU loading and perhaps see a performance boost to the algorithm if it was bottle-necked on parallel processing power. But this will greatly depend on the usecase, such as image size, number of images, etc.

    Regards,

    Erick

  • 0 in reply to Erick Narvaez
    Expert 1501 points

    Hi，

        This reply is helpful to us, thank you!

         Regards,

         Kong