Deployment using c++

S.sarkar · April 25, 2023, 10:51am

I am deploying my model on my Mac using C++. After doing all the processes why only 11 parameters are showing as output? I have used a dataset that contains approx 1000 images. I have labeled the images using a bounding box. Each T-shirt image has 12 labeled points ( 2 HSP, 2 LSP, 4AP, 2 UP, 2WP). Why it is showing like this please help me out.

louis · April 25, 2023, 12:24pm

Hello @S.sarkar,

The model here detected 11 boxes (2 HSP, 1 LSP, 3 AP, 3 WP and 2 UP).
If you want the center of the objects, just do x+4 and y+4; 4 being the height/2 and width/2 of the object.

Best,

Louis

S.sarkar · April 25, 2023, 12:31pm

please tell me how can I extract the coordinates as my project is about "automatic garment measurement’. I want to make something, where from clicking an image my desktop should automatically tell the garment’s measurement. how can I extract the coordinates as I have trained the model on edge impulse and also run the code on Mac. what will be the next step?

louis · April 25, 2023, 12:47pm

Hello @S.sarkar,

Here is the place where you can grab the coordinates with our C++ SDK for Linux:

github.com

edgeimpulse/example-standalone-inferencing-linux/blob/master/source/camera.cpp#L149


      
          #if EI_CLASSIFIER_OBJECT_DETECTION == 1
              printf("Classification result (%d ms.):\n", result.timing.dsp + result.timing.classification);
              bool found_bb = false;
              for (size_t ix = 0; ix < result.bounding_boxes_count; ix++) {
                  auto bb = result.bounding_boxes[ix];
                  if (bb.value == 0) {
                      continue;
                  }
          
          
        found_bb = true;
                  printf("    %s (%f) [ x: %u, y: %u, width: %u, height: %u ]\n", bb.label, bb.value, bb.x, bb.y, bb.width, bb.height);
              }
          
          
    if (!found_bb) {
                  printf("    no objects found\n");
              }
          #else
              printf("%d ms. ", result.timing.dsp + result.timing.classification);
              for (size_t ix = 0; ix < EI_CLASSIFIER_LABEL_COUNT; ix++) {
                  printf("%s: %.05f", result.classification[ix].label, result.classification[ix].value);
                  if (ix != EI_CLASSIFIER_LABEL_COUNT - 1) {

You might want to have a look at our Linux CLI (which also works on Mac):

or our Linux Python SDK

See also for advanced usage using the Python SDK:

Best,

Louis