“ArrangementNet: Learning Scene Arrangements for Vectorized Indoor Scene Modeling” by Huang, Zhang, Duan, Zhang, Guo, et al. …
Conference:
Type:
Title:
- ArrangementNet: Learning Scene Arrangements for Vectorized Indoor Scene Modeling
Session/Category Title: Geometric Abstractions: Not Just for Cubists
Presenter(s)/Author(s):
Moderator(s):
Abstract:
We present a novel vectorized indoor modeling approach that converts point clouds into building information models (BIM) with concise and semantically segmented polygonal meshes. We propose ArrangementNet to estimate scene arrangements and faithfully model challenging scenes from laser scans or multiview stereo. We show significant improvement in BIM model reconstruction.
