According to the process of function parse in class HeatmapParser, there has three outputs of the function top_k:
tag_k has the shape of _(num_images, num_joints, max_numpeople, 2);
loc_k has the same shape with tag_k;
val_k has the shape of _(num_images, num_joints, max_numpeople).
The outputs of the function top_k will be the inputs of the function match_by_tag directly, however, according to the process of function match_by_tag, the first dimension of the three outputs has become as _numjoints rather than _numimages. The evidence is as follows:
for i in range(params.num_joints):
idx = params.joint_order[i]
tags = tag_k[idx]
According to the process of function parse in class HeatmapParser, there has three outputs of the function top_k: tag_k has the shape of _(num_images, num_joints, max_numpeople, 2); loc_k has the same shape with tag_k; val_k has the shape of _(num_images, num_joints, max_numpeople). The outputs of the function top_k will be the inputs of the function match_by_tag directly, however, according to the process of function match_by_tag, the first dimension of the three outputs has become as _numjoints rather than _numimages. The evidence is as follows: