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.

TDA4AEN-Q1: Does TI DMPAC have random traits?

Part Number: TDA4AEN-Q1

Dear, TI experts.

I have a small question regarding the TI DMPAC specification.

I'm currently developing the MCAM (Mobis Front Camera) product using the TDA4AEN, and I'm curious about some characteristics of the TI DMPAC. (I'm now working on S/W algorithm implementation using TI optical flow.)

I hadn't expected the DOF (Dense Optical Flow) output from DMPAC to vary using the same sequence of image data, but recently I started to suspect there there might be some randomness, and one of my colleauges who has more experience than I do, mentioned that the HWA can show random behavior. Is that ture?

I'm looking forward to your answers. Thank you.

Regards,
Minsun

  • Hi Minsun,

    No DOF output does not have randomness, it detects and outputs the flow vector based on the input images and configuration, so as long as input is correct and also configuration, it can output correct.. 

    So where do you see randomness? Can you please elaborate on this? 

    Regards,

    Brijesh

  • Hi Brijesh

    Thank you for the answer.

    I think it would be better to change the expression from "randomness" to "not determinisic".

    The optical flow output from DMPAC is correct, but what I mean is that the optical flow output are not the same even when running multiple times with the same images.

    Is it possible?

    Regards,

    Minsun

  • Hi Minsun,

    no, if the input is same and configuration remains same, it will generate same flow vector output.. 

    Regards,

    Brijesh