AI Paper Atlas

Research map / 3D vision / Face and shape

Learning Generative Models of Textured 3D Meshes From Real-World Images

Dario Pavllo, Jonas Kohler, Thomas Hofmann, Aurelien Lucchi

ICCV 2021 · 3D vision › Face and shape · approach: Generative textures for 3D humans · 37 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
add semantic template inference after silhouette optimization (ablation)pose estimation, single template · Bird (CUB) poseGD(1): 0.47 -> 0.29Improved
add semantic template inference after silhouette optimization (ablation)pose estimation, single template · Car (P3D) poseGD(1): 0.12 -> 0.11No clear effect
add semantic template inference after silhouette optimization (ablation)pose estimation, multiple templates · Car (P3D) poseGD(1): 0.10 -> 0.06Improved
use multiple mesh templates instead of one (ablation)pose estimation with semantics · Bird (CUB) poseGD(1): 0.29 -> 0.32Hurt
use multiple mesh templates instead of one (ablation)pose estimation with semantics · Car (P3D) poseGD(1): 0.11 -> 0.06Improved
repeat ambiguity detection and semantic inference a second time (ablation)pose estimation, single template · Bird (CUB) poseGD(1): 0.29 -> 0.29No clear effect
use full pipeline with semantics instead of silhouette-only poses (ablation)generator, single template · CUB bird FIDFID: 73.67 -> 55.75Improved
use full pipeline with semantics instead of silhouette-only poses (ablation)generator, single template · P3D car FIDFID: 38.16 -> 36.52Improved
use full pipeline with semantics instead of silhouette-only poses (ablation)generator, single template · P3D airplane FIDFID: 100.5 -> 81.28Improved
use multiple templates instead of one (ablation)generator with semantics · CUB bird FIDFID: 55.75 -> 56.54No clear effect
train one conditional generator for all classes (setting B) instead of per-class models (ablation)generator, single template · ImageNet airplane FIDFID: 77.84 -> 46.46Improved
train one conditional generator for all classes (setting B) instead of per-class models (ablation)generator, single template · ImageNet car FIDFID: 30.73 -> 39.68Hurt

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