• State Resolved
  • Locked Locked
  • Replies 4 replies
  • Answers 1 answer
  • Subscribers 98 subscribers
  • Views 862 views
  • Users 0 members are here
Support feedback
Options
Options
Related

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.

RTOS / TDA2： How to run TIDL_OD usecase?

Allen Cheng
Intellectual 620 points
Part Number: TDA2

Tool/software: TI-RTOS

Hi,

I try to run TIDL_OD usecase

but I got fail

Can you tell anything to me?

How do I do? Can you provide guide for our reference? 

  • 0
    TI__Expert 3970 points

    VisionSDK_UserGuide_TI_DeepLearning.pdf

    describes the section
    5 Build and Run TIDL Object detect use case

    Before running the Vision SDK binary,
     Make sure the following files are present in the MMC/SD card:
    o TIDL OD Network file (NET_OD.bin)
    o TIDL OD Parameter file (PRM_OD.bin)
    o TIDL OD Usecase input data file (inData_OD)
    o TIDL OD Usecase input header file (inHeader_OD)

    The error is seen as the files are not copied into SD card .

    Please copy the input files and rerun the usecase .

    Regards
    Chetan.M

  • 0 in reply to Chetan Matad
    TI__Expert 3970 points
    Allen,

    Are you able to test with the given changes ,
    If not please mention what is the error you are facing.

    Regards
    Chetan.M
  • 0 in reply to Chetan Matad
    Intellectual 620 points
    Oh ,thanks

    I can run the TIDL_OD use case.

    but,

    Can we use the frame by camera capture for input data in TIDL_OD use case?

    If yes,

    How do I do? Can you provide any guide for our reference?

    If no

    I want to run TIDL_OD Demo with OV Sensors.

    How can I to do this?
  • 0 in reply to Allen Cheng
    TI__Expert 3970 points
    Hi Allen ,

    Please close this thread and start another thread for running TIDL on live camera,

    You can always modify the use-case to remove the null src and have capture source and set the params accordingly

    I am not sure how to train the model and other details

    Regards
    Chetan.M