“…"Landmarks ": [267,137,261,304,214,154,313,154,239,202,294,192,227,261,295,258,247,286,273,291] "links " : {[ [9,10], [7,8], [5,6], [3,4], [1,2]]},…]} Using this information, heatmaps and vector field maps for each image are labeled for network training.…”