AI Paper Atlas

Research map / 3D vision / Face and shape

In-Hand 3D Object Scanning From an RGB Sequence

Shreyas Hampali, Tomas Hodan, Luan Tran, Lingni Ma, Cem Keskin, Vincent Lepetit

CVPR 2023 · 3D vision › Face and shape · approach: Single-view 3D reconstruction · 17 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
Estimate object shape and camera trajectory jointly without supplied posesIn-Hand 3D Object Scanning · HO-3D, averageRMSE Hausdorff distance (mm): 13.41 -> 4.65Improved
Estimate shape and poses from RGB rather than use the single-image Ye et al. methodHO-3D · HO-3D, averageRMSE Hausdorff distance (mm): 6.76 -> 4.65Improved
Estimate poses instead of using ground-truth posesOurs versus UNISURF · HO-3D, averagePSNR: 31.02 -> 31.01No clear effect
Estimate poses instead of using ground-truth posesOurs versus UNISURF · HO-3D, averageSSIM: 0.75 -> 0.77Improved
Estimate poses instead of using ground-truth posesOurs versus UNISURF · HO-3D, averageLPIPS: 0.34 -> 0.32Improved
Add synthetic-depth loss (ablation)In-Hand 3D Object Scanning · HO-3D MDF14, largest segmentATE AUC: 4.9 -> 5.8Improved
Add optical-flow loss (ablation)In-Hand 3D Object Scanning · HO-3D MDF14, largest segmentATE AUC: 4.8 -> 5.8Improved
Add shape-regularization loss (ablation)In-Hand 3D Object Scanning · HO-3D MDF14, largest segmentATE AUC: 1.1 -> 5.8Improved
Add synthetic-depth loss (ablation)In-Hand 3D Object Scanning · HO-3D SM2, largest segmentATE AUC: 1.0 -> 8.1Improved
Add optical-flow loss (ablation)In-Hand 3D Object Scanning · HO-3D SM2, largest segmentATE AUC: 2.9 -> 8.1Improved
Add shape-regularization loss (ablation)In-Hand 3D Object Scanning · HO-3D SM2, largest segmentATE AUC: 0.5 -> 8.1Improved
Split the sequence into overlapping segments, initialize from segment pose estimates, then optimize globallyIn-Hand 3D Object Scanning · In-hand RGB sequence reconstructionconvergenceImproved

Earlier methods this paper compares with

Named baselines from earlier accepted papers in this paper's results tables: better than 0 of 1 on most shared results.

Matched by method name in results tables read by AI. Authors' own tables tend to favour the newer method, and a comparison can be missed when a method is renamed.

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