• State Resolved
  • Locked Locked
  • Replies 4 replies
  • Answers 1 answer
  • Subscribers 29 subscribers
  • Views 356 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.

IWR1642BOOST: The value of "mInde" when tracking is done in PC mode is wrong

koji fukumoto
Intellectual 940 points
Part Number: IWR1642BOOST

Hello.

When tracking is done with "tm_play", "mIndex" is returned as the result of the step process.
When tracking in PC mode, mIndex value of tracking start frame is wrong.

---------------------------------
1. Start tm_play
2. fHist file select the following
C:\ti\mmwave_industrial_toolbox_2_4_0\labs\lab0013_traffic_monitoring_16xx\gui\sample_fhist\fHistRT_0001.mat
3. Select "PC" in Replay mode
4. Select the following for the config file
C:\ti\mmwave_industrial_toolbox_2_4_0\labs\lab0013_traffic_monitoring_16xx\chirp_configs\mmw_tm_demo_ph2.cfg
5. Enter "-5, 0, 4" in "Other Mandatory Parameters"
6. Press the start button
---------------------------------

Tracking is done for the first time in the 29th frame.
At this time, mIndex returned from step processing is [0, 0, 0, 0, 0].
The mIndex contains the tracking number, but the value 0 is invalid (because it starts with 1).

From the 30th frame, 1 is entered for mIndex.

I think there may be a problem with trackModule.m or trackUnit.m.

I want to acquire mIndex correctly.
How can I fix it?

Thanks.

  • Former Member
    0 Former Member

    Hello Koji,

    In C code the indexing in gtrack starts at 0 while it is 1 in MATLAB so there's a translation in the tm_play.

    Can you give details on how this discrepancy affects your tracker results?

    Amanda

  • 0 in reply to Former Member
    Intellectual 940 points

    Hello, Amanda.
    Thanks for the reply.

    When tm_play is tracked in PC mode, "gtrack algorithm created in MATLAB language" is executed.
    The source code is "trackModule.m" or "trackUnit.m".
    This algorithm numbers the tracking ID from 1.
    (On the other hand, the gtrack algorithm created in C language is numbered from 0)

    Please see line 308 of tm_play.m.
    Step 1 is subtracted from mIndex obtained by processing.
    This process is executed only in PCMatrab mode.

    mIndex = mIndex - 1;

    Next, look at line 343.
    This time, 1 is added to mIndex.
    This process is executed in both PCMatlab mode and PCMex mode.

    mIndex = mIndex + 1;

    With these calculations, mIndex is adjusted as follows.
    · In PCMatlab mode, adopt the mIndex returned by gtrack as it is (since gtrack and tm_play all start with index 1)
    · In the case of PCMex mode, it adopts mIndex returned by gtrack plus 1 (since the index of gtrack starts from 0, the index of tm_play starts from 1)

    The problem is the mIndex of the tracking start frame in PCMatlab mode.
    The mIndex returned by gtrack is [0, 0, 0, 0, 0].
    The index of gtrack in PCMatlab starts with 1, but returning 0 is strange.

    In tm_play, this problem hardly surfaced.
    There is little processing to refer to mIndex.

    I am creating a tracking application with reference to tm_play.
    In that application we often refer to mIndex.
    Therefore, I want to solve this problem.

    Thanks.

  • Former Member
    0 Former Member in reply to koji fukumoto
    Hello Koji,

    As you pointed out this is extraneous code that is not the core of tm_play. The source code for gtrack can be modified and recompiled if you prefer mIndex to behave differently.

    Thanks,
    Amanda
  • 0 in reply to Former Member
    Intellectual 940 points

    Hi Amanda.
    thank you for your reply.

    I'd like to know how to correct mIndex in the correct way.
    But does not TI know how to fix it?
    It is very disappointing.

    Thanks.