AI Paper Atlas

Research map / 3D vision / Face and shape

Rasterized Edge Gradients: Handling Discontinuities Differentially

Stanislav Pidhorskyi, Tomas Simon, Gabriel Schwartz, He Wen, Yaser Sheikh, Jason Saragih

ECCV 2024 · 3D vision › Face and shape · approach: Neural implicit surface 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 discontinuity (edge) gradients instead of continuous-only gradients (ablation)mesh inverse rendering · Blender synthetic (Lego)PSNR: 16.342 -> 29.667Improved
replace nvdiffrast antialiasing gradients with EdgeGradmesh inverse rendering · Blender synthetic (Chair)PSNR: 29.804 -> 32.981Improved
replace nvdiffrast antialiasing gradients with EdgeGradmesh inverse rendering · Blender synthetic (Ficus)PSNR: 24.964 -> 26.559Improved
add handling of geometry intersections (ablation)EdgeGrad · Blender synthetic (8 scenes)PSNR: Lego 29.574 -> 29.667; Ficus 26.781 -> 26.559Mixed
add 2D segmentation mask supervision (ablation)dynamic head avatar fitting · dynamic head scenesqualitativeImproved
use EdgeGrad instead of nvdiffrast gradientsgradient computation at intersecting geometry · forward gradient test scenesmatch to finite differencesImproved
use EdgeGrad instead of Mitsuba on small imagesgradient accuracy · gradient error by image sizegradient errorHurt

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