AI Paper Atlas

Research map / 3D vision / Face and shape

Learning Mesh Representations via Binary Space Partitioning Tree Networks

Zhiqin Chen, Andrea Tagliasacchi, Hao Zhang

TPAMI 2021 · 3D vision › Face and shape · approach: Primitive-based shape abstraction · 13 citations (OpenAlex)

Read the paperCheck your idea against 137,217 papers

AI-extracted. The measured effects below were read by AI from the paper's full text (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.

Measured effects reported in the paper

What was changedWhereMeasuredEffect
use BSP-Net instead of BAEshape decomposition · ShapeNet auto-encodingChamfer Distance: 1.592 -> 0.447Improved
use BSP-Net instead of BAEshape decomposition · ShapeNet auto-encodingLight Field Distance: 4587.34 -> 2019.26Improved
add overlap loss (ablation)BSP-Net · ShapeNet auto-encodingChamfer Distance: 0.447 -> 0.448No clear effect
add overlap loss (ablation)BSP-Net · ShapeNet Part segmentationmean IoU: 79.3 -> 79.8Improved
use BSP-Net instead of BAE (one-shot)segmentation · ShapeNet Partmean IoU: 59.8 -> 79.3Improved
add soft threshold loss L*_T (ablation)BSP-Net · shape auto-encodingChamfer Distance (before): 0.743 -> 0.735Improved
add L*_T and overlap loss (ablation)BSP-Net · shape auto-encodingChamfer Distance (before): 0.743 -> 0.724Improved
add L*_T and overlap loss (ablation)BSP-Net · single view reconstructionChamfer Distance (before): 1.432 -> 1.453No clear effect
vary threshold lambda (ablation)BSP-Net · shape auto-encodingChamfer Distance: 0.743 -> 0.758No clear effect
use BSP-Net instead of Atlas25 for sharp edgessingle view reconstruction · ShapeNet SVREdge Chamfer Distance: 2.069 -> 0.743Improved
use BSP-Net instead of Atlas25single view reconstruction · ShapeNet SVRChamfer Distance: 1.170 -> 1.432Hurt
use BSP-Net instead of IM-NET256 for mesh sizesingle view reconstruction · ShapeNet SVRvertex count: 82965 -> 1073Improved

Closest papers by meaning

Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.

Open scime draft

Get the next scime Dispatch and a monthly field update

Free. We send a confirmation link first; every email has a one-click unsubscribe. Privacy