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.

TDA2: tda2 TIDL SSD

Part Number: TDA2

Dear TI,

            A strange bug comes, I find that  after I run the TIDL demo  , the output detection ROI coordinate is not correct!
            As you see below,I printed the detected rectangle coordinate ,then check the output value of the rectangle coordinate,nearly all of those are not accurate location .
             Is there any constant value needed be substracted from the variable pObjectDateDesc->ypos and  pObjectDateDesc->xpos  ,even pObjectDateDesc->objHeight and pObjectDateDesc->objWith needed be multiplied a constan coefficient?

            In fact, this page that you answer me  is not the right information I want。
           Could you please chech the file "vision_sdk\apps\src\rtos\alg_plugins\objectdetection\objectDrawLink_algPlugin.c" ? In this file, I print the variable pObjectDateDesc->ypos and  pObjectDateDesc->xpos  ,even pObjectDateDesc->objHeight and pObjectDateDesc->objWith when detecting the vehicle, yet the result rectangle is far from the real vehicle objects in the image.
           Is there any math relation of the results coordinate?


           Looking forward to your kind reply, tks dear.