search

DirtyHarryLYL / DJ-RN

As a part of HAKE project (HAKE-3D). Code for our CVPR2020 paper "Detailed 2D-3D Joint Representation for Human-Object Interaction".
Apache License 2.0
100 stars 13 forks source link

The visualized attention #60

Open mary-arch opened 3 years ago

mary-arch commented 3 years ago

Hello How extract the visualized attention on of 3d Human(figure 8 in your paper)?

Foruck commented 3 years ago

You could try to save these two predictions for visualization. https://github.com/DirtyHarryLYL/DJ-RN/blob/83c3d0f16a4a3b55545e96a8350397f6adb1ef91/lib/networks/DJR.py#L550-L551

mary-arch commented 3 years ago

Hello when I print A_3D.shape, I got size (17, 17) for example in image HICO_train2015_00000148. But in your paper you visualized a mesh of human. how you get from (17,17) to point clouds and visualized mesh?

mary-arch commented 3 years ago

You answer the first question and do not answer the rest of the questions. Why?

yangyuan333 commented 2 years ago

Have you solved this problem? I also encountered the same problem, how to visualize the 2D attention map