search

Jieqianyu / SGN

Implementation of IEEE TIP 2024 paper - "Camera-based Semantic Scene Completion with Sparse Guidance Network"
Other
24 stars 3 forks source link

About FOV mask shape in paper. #2

Closed hongxiaoy closed 8 months ago

hongxiaoy commented 8 months ago
image

Hi, thanks for your nice work. I have a question when I read the paper, just as the image shown, in my opinion this mask should be a 3D mask for voxels, with voxels projected outside the image be False, and projected in the image be True. I don't know if my understanding has some bias. Hope your reply, thanks~

Jieqianyu commented 8 months ago

Thank you for your interest in our work. You are right, it is indeed a 3D mask for voxels. We are sorry that this typo error confused you. Thank you again for reading carefully and helping us correct our mistakes!