AI Paper Atlas

Research map / Point clouds / Shape and surface completion

Intrinsic Point Cloud Interpolation via Dual Latent Space Navigation

Marie-Julie Rakotosaona, Maks Ovsjanikov

ECCV 2020 · Point clouds › Shape and surface completion · approach: Unsupervised shape correspondence

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
interpolate in edge-length latent space and decode with shape decoder (dual latent space) instead of linear interpolation in PointNet AE latent spacePointNet AE · DFAUST interpolationedge length variance: 0.3510 -> 0.2311Improved
dual latent space instead of PointNet AE on animalsPointNet AE · SMAL interpolationedge length variance: 2.068 -> 1.538Improved
dual latent space in unsupervised settingPointNet AE (unsupervised) · Unsupervised DFAUST interpolationedge length variance: 0.597 -> 0.398Improved
add mapping networks to shape AE for reconstructionPointNet AE · DFAUST reconstructionedge length error (1e-5): 3.023 -> 1.641Improved
add mapping networks to shape AE for reconstructionPointNet AE · DFAUST reconstructionpoint cloud error (1e-4): 2.120 -> 2.572Hurt
add mapping networks for reconstruction on unseen datasetShape AE · SCAPE reconstructionChamfer distance (1e-3): 4.703 -> 4.135Improved
add cycle-consistency losses instead of direct reconstruction losses (ablation)mapping networks · DFAUST reconstructionedge length error: 0.1019 -> 1.641*10^{-5}Improved
add L_map1 (rotation-invariant shape loss) to L_map2, L_map3 (ablation)mapping networks · DFAUST reconstructionpoint cloud error PC (1e-4): 14.816 -> 2.572Improved
add L_map2 to L_map1, L_map3 (ablation)mapping networks · DFAUST reconstructionedge length error EL (1e-5): 2.301 -> 1.641Improved
add L_map1 to L_map2, L_map3 effect on edge error (ablation)mapping networks · DFAUST reconstructionedge length error EL (1e-5): 1.595 -> 1.641No clear effect
add linearity regularization L_lin to edge auto-encoder (ablation)edge auto-encoder · edge AE latent interpolationedge length variance: 1.777 -> 0.199Improved
add linearity regularization L_lin to edge auto-encoder (ablation)full method · DFAUST interpolationedge length variance: 0.245 -> 0.230Improved

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