AI Paper Atlas

Research map / Point clouds / Shape and surface completion

Robust and Accurate Superquadric Recovery: A Probabilistic Approach

Weixiao Liu, Yuwei Wu, Sipu Ruan, Gregory S. Chirikjian

CVPR 2022 · Point clouds › Shape and surface completion · approach: Unsupervised shape correspondence · 48 citations (OpenAlex)

Check 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 EMS with geometry-guided switching instead of baseline fitting methodsSingle superquadric recovery · Synthetic partial point cloudsaverage fitting errorImproved
Use EMS with geometry-guided switching instead of baseline fitting methodsSingle superquadric recovery · Synthetic partial point cloudsglobal optimality rateImproved
Add the S-step geometry-guided switching strategy (ablation)EMS · Synthetic partial point cloudsfitting errorImproved
Use probabilistic outlier inference with a uniform outlier componentSingle superquadric recovery · Synthetic point clouds with outliersaverage fitting errorImproved
Use posterior probabilistic outlier inference instead of heuristic outlier rejectionSingle superquadric recovery · Clean synthetic point cloudsaverage fitting errorImproved
Add Gaussian noise to the input point cloudAll evaluated recovery methods · Synthetic noisy point cloudsfitting errorHurt
Use EMS instead of baseline methodsSingle superquadric recovery · Synthetic noisy point cloudsrobustnessImproved
Use EMS instead of baseline methodsSingle superquadric recovery · KIT ObjectModelsaverage fitting errorImproved
Use EMS instead of baseline methodsSingle superquadric recovery · BigBIRDaverage fitting errorImproved
Use EMS instead of baseline methodsSingle superquadric recovery · BigBIRDruntimeImproved
Recover multiple superquadrics hierarchically from points identified as outliersEMS multi-superquadric extension · Complex KIT ObjectModelsrecovery of major superquadric-like partsImproved

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