what is the purpose of zarray_get(detections_, i, &detection) and how set H Matrix

[Wuxinxiaoshifu]

there is an H matrix in addImagePoints(detection, standaloneTagImagePoints) , i'm wondering how H is get int your function?

some details: 1.homography_project(detection->H, tag_x[i], tag_y[i], &im_x, &im_y); 2. // assuming that the projection matrix is: // [ fx 0 cx 0 ] // [ 0 fy cy 0 ] // [ 0 0 1 0 ] // // And that the homography is equal to the projection matrix times the model matrix, // recover the model matrix (which is returned). Note that the third column of the model // matrix is missing in the expresison below, reflecting the fact that the homography assumes // all points are at z=0 (i.e., planar) and that the element of z is thus omitted. // (3x1 instead of 4x1). // // [ fx 0 cx 0 ] [ R00 R01 TX ] [ H00 H01 H02 ] // [ 0 fy cy 0 ] [ R10 R11 TY ] = [ H10 H11 H12 ] // [ 0 0 1 0 ] [ R20 R21 TZ ] = [ H20 H21 H22 ] // [ 0 0 1 ] // // fxR00 + cxR20 = H00 (note, H only known up to scale; some additional adjustments required; see code.) // fxR01 + cxR21 = H01 // fxTX + cxTZ = H02 // fyR10 + cyR20 = H10 // fyR11 + cyR21 = H11 // fyTY + cyTZ = H12 // R20 = H20 // R21 = H21 // TZ = H22

[Wuxinxiaoshifu]

i also wondering whether the H is known or unknown before running the homography_project function?

[Wuxinxiaoshifu]

in others words, the function of apriltag_detector_detect calculate the H?

[wxmerkt]

Has your question been resolved? I am not sure I am following based on the pointers given so far