AI Paper Atlas

Research map / 3D vision / Face and shape

The Phong Surface: Efficient 3D Model Fitting using Lifted Optimization

Jingjing Shen, Thomas J. Cashman, Qi Ye, Tim Hutton, Toby Sharp, Federica Bogo, Andrew Fitzgibbon, Jamie Shotton

ECCV 2020 · 3D vision › Face and shape · approach: Pixel-aligned implicit human and hand reconstruction

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 data normal term (ablation)Phong surface · Ellipsoid rigid pose alignment (lifted)avg rotation error at 10 iters: 23.24 -> 8.13Improved
add data normal term (ablation)Phong surface · Ellipsoid rigid pose alignment (lifted)avg rotation error at 50 iters: 3.54 -> 0.99Improved
add data normal term (ablation)Loop subdivision surface · Ellipsoid rigid pose alignment (lifted)avg rotation error at 10 iters: 14.81 -> 9.89Improved
add data normal term (ablation)Loop subdivision surface · Ellipsoid rigid pose alignment (lifted)avg rotation error at 50 iters: 2.57 -> 1.21Improved
add data normal term (ablation)Triangular mesh · Ellipsoid rigid pose alignment (lifted)avg rotation error at 10 iters: 23.24 -> 17.47Improved
add data normal term (ablation)Triangular mesh · Ellipsoid rigid pose alignment (lifted)avg rotation error at 50 iters: 3.54 -> 11.07Hurt
use Phong surface instead of triangular meshlifted optimization · Ellipsoid rigid pose alignment (lifted)avg rotation error at 50 iters: 11.07 -> 0.99Improved
use Phong surface instead of Loop subdivisionlifted optimization · Ellipsoid rigid pose alignment (lifted)avg rotation error at 50 iters: 1.21 -> 0.99Improved
use lifted optimization instead of ICPPhong surface · Ellipsoid rigid pose alignmentiterations to reach error <10 deg: 30 iterations -> 8 itersImproved
use Phong surface instead of Loop subdivisionlifted optimization on HoloLens 2 · Hand tracking, HoloLens 2 depth dataevaluation speed: twice as fastImproved
use Phong surface instead of Loop subdivisionlifted hand tracker · Dexter, per-frame fitting at 50 fpsiterations affordable: 4 iterations -> 6 iterationsImproved
use Phong surface instead of Loop subdivisionlifted hand tracker · Dexter, near 80% accuracyspeed: 20% fasterImproved

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