search

705062791 / PGBIG

CVPR 2022
41 stars 8 forks source link

Question about Ignored_dims #1

Open Xiyan-Xu opened 2 years ago

Xiyan-Xu commented 2 years ago

I notice that in line 273~276 of the file "main_cmu_3d.py", _p3d_out4 is treated as follow:

        p3d_out_4 = p3d_h36.clone()[:, in_n:in_n + out_n]

        p3d_out_4[:, :, dim_used] = p3d_out_all_4[:, seq_in:]

        p3d_out_4[:, :, index_to_ignore] = p3d_out_4[:, :, index_to_equal]

        p3d_out_4 = p3d_out_4.reshape([-1, out_n, all_dim//3, 3])

Will this result in some joints being counted multiple times when calculating MPJPE?

Thanks

705062791 commented 2 years ago

Thank you for your interest in our work. This is a general approach. Because there are some overlapping points in human joints, these points are usually removed during prediction. However, when the error is calculated, these points are re-added to ensure the integrity of the pose. As far as I know, all existing methods do so.