This is an official implementation of our CVPR 2020 paper "HigherHRNet: Scale-Aware Representation Learning for Bottom-Up Human Pose Estimation" (https://arxiv.org/abs/1908.10357)
In group.py, inside the function 'parse', the variable 'ans' is a list of length "batch_size", each element of it is a numpy ndarray which has shape of (num_people, 17, 5) when FLIP_TEST is True. So my question is about the code as follows:
scores = [i[:, 2].mean() for i in ans[0]]
What's the meaning of ans[0]? The first image in a minibatch? What's the meaning of i[:, 2]? I think i[:, 2] has the shape of (num_people, 5), I just wonder the meaning of i[:, 2].mean()...
In group.py, inside the function 'parse', the variable 'ans' is a list of length "batch_size", each element of it is a numpy ndarray which has shape of (num_people, 17, 5) when FLIP_TEST is True. So my question is about the code as follows:
scores = [i[:, 2].mean() for i in ans[0]]What's the meaning of ans[0]? The first image in a minibatch? What's the meaning of i[:, 2]? I think i[:, 2] has the shape of (num_people, 5), I just wonder the meaning of i[:, 2].mean()...