Not understanding what bbox is

[RafaRuiz]

Hello, I've seen an example, for example the examples/video_input.cpp and I can see this when passing from cv::Rect (face_rect) to bbox:
bbox[0] = faces[i].x;
bbox[1] = faces[i].y;
bbox[2] = faces[i].x + faces[i].width;
bbox[3] = faces[i].y;
bbox[4] = faces[i].x + faces[i].width;
bbox[5] = faces[i].y + faces[i].height;
bbox[6] = faces[i].x;
bbox[7] = faces[i].y + faces[i].height;

the parameter 0 and 6 is the same, the parameter 1 and 3 is the same, 2 and 4, 5 and 7.

Why?

[uricamic]

Hi @RafaRuiz,

CLandmark supports a general face box given by coordinates of all its corners. However, most of the face detectors (including the OpenCV one used in the example) provide only axis aligned face boxes. That's why some of the coordinates seem to be duplicates.

[RafaRuiz]

Fair enough, thank you very much!

[uricamic]

No problem, I am glad it helped.