AI Paper Atlas

Research map / Point clouds / Shape and surface completion

Detail Preserved Point Cloud Completion via Separated Feature Aggregation

Wenxiao Zhang, Qingan Yan, Chunxia Xiao

ECCV 2020 · Point clouds › Shape and surface completion · approach: Point cloud completion networks · 171 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
add RFA separated feature aggregation (ablation)NWoSFA · ShapeNet known categoriesChamfer distance: 8.74 -> 8.06Improved
add GLFA separated feature aggregation (ablation)NWoSFA · ShapeNet known categoriesChamfer distance: 8.74 -> 8.14Improved
add RFA aggregation (ablation)NWoSFA · ShapeNet novel categoriesChamfer distance: 10.02 -> 10.08No clear effect
add GLFA aggregation (ablation)NWoSFA · ShapeNet novel categoriesChamfer distance: 10.02 -> 9.98Improved
add refinement component (ablation)NSFA-RFA · ShapeNet known categoriesChamfer distance: 8.65 -> 8.06Improved
add refinement component (ablation)NSFA-GLFA · ShapeNet known categoriesChamfer distance: 9.38 -> 8.14Improved
add refinement component (ablation)NSFA-RFA · ShapeNet novel categoriesChamfer distance: 10.09 -> 10.08No clear effect
add refinement component (ablation)NSFA-GLFA · ShapeNet novel categoriesChamfer distance: 10.70 -> 9.98Improved
use RFA instead of GLFA aggregationNSFA · ShapeNet known categoriesChamfer distance: 8.14 -> 8.06Improved
increase feature extraction levels from 5 to 6 (ablation)NSFA-GLFA · ShapeNet known categoriesChamfer distanceHurt
set known:missing mixture ratio to 1:3 or 0:4 instead of near 1:1 (ablation)NSFA · ShapeNet known categoriesChamfer distanceHurt
proposed vs PCN on dissimilar novel categoriesNSFA-RFA · ShapeNet novel categories (dissimilar)Chamfer distance: 12.90 -> 9.20Improved

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