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.

Résolution facedetect demo

Hello !!

I would know if we can do the HD acquisition with facedetect demo. As mentioned in the demo presentation : " The demo also supports D1(720x480)".

 

Thank you

Mus

  • Hi Mustapha,

    Thanks for your post.

    You are right. It should support D1 resolution.

    Thanks & regards,

    SIvaraj K

    -------------------------------------------------------------------------------------------------------
    Please click the Verify Answer button on this post if it answers your question.
    -------------------------------------------------------------------------------------------------------

  • Thank you !!

    What about the quality of the display video of the face detectdemo ? It is a good quality. What we can do to make it better?

    Mus

  • Hi Mustapha,

    The quality all depends on the input resolution and obviously, you will have signal degration in video output but quite reasonalbel for display. If you still need output quality better, you need to apply video post processing software algorithms to append over the existing code and configure over VGA display.

    Thanks & regards,

    Sivaraj K

    -------------------------------------------------------------------------------------------------------
    Please click the Verify Answer button on this post if it answers your question.
    -------------------------------------------------------------------------------------------------------

  • Thank you !!

    Could you give a link or some name of functions that do that. I'm beginner in video processing

    Mus

  • Hi Mustapha,

    There are lot many robust motion compensation algorithms available as open source to improve video quality based out of video resoultion standards. Please google it, you will find many resources on the same. To nullify the residual error between the predicted video frames thru. motion vectors and actual video (raw) frames, there are lot many motion vector estimation algorithms across Intra-prediction (I-frames) and Inter-prediction ( Forward P-frames & bidirectional B-frames prediction) are available as open source. By this, you could understand, how to reduce the residual error and to improve video quality on standard output resoultions.

    Thanks & regards,

    Sivaraj K

    -------------------------------------------------------------------------------------------------------

    Please click the Verify Answer button on this post if it answers your question

    -------------------------------------------------------------------------------------------------------

  • Hi Sivaraj

    Have you some intresting link?

    Tnak you !!

    Mus

  • Hi Mustapha,

    Please google it to find out the open source links and you have lot many sources available. Actually we are not supposed to provide links other than TI's proprietary links. We apologise for the same

    Kinldy close the thread, if you do not have any further questions on this.

    -------------------------------------------------------------------------------------------------------
    Please click the Verify Answer button on this post if it answers your question.
    -------------------------------------------------------------------------------------------------------