AI Paper Atlas

Research map / 3D vision / Face and shape

Preface: A Data-driven Volumetric Prior for Few-shot Ultra High-resolution Face Synthesis

Marcel C. Bühler, Kripasindhu Sarkar, Tanmay Shah, Gengyan Li, Daoye Wang, Leonhard Helminger, Sergio Orts-Escolano, Dmitry Lagun, Otmar Hilliges, Thabo Beeler, Abhimitra Meka

ICCV 2023 · 3D vision › Face and shape · approach: Face reflectance and relightable rendering · 23 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
initialise target latent by inversion rather than zeros, noise, mean or neighbour latents (ablation)Preface model fitting · Studio dataset, 2 views, 1KPSNR: 24.66 -> 25.69Improved
initialise from furthest training identity instead of inversion (ablation)Preface model fitting · Studio dataset, 2 views, 1KPSNR: 25.69 -> 23.91Hurt
add view-branch weight regulariser L_v (ablation)Preface fine-tuning · Studio dataset, 2 views, 1KPSNR: 23.91 -> 25.53Improved
add normal consistency loss L_normal (ablation)Preface fine-tuning · Studio dataset, 2 views, 1KPSNR: 25.53 -> 25.69Improved
increase number of fitting views from 2 to 7 (ablation)Preface · Studio dataset, 1KPSNR: 25.69 -> 29.24Improved
train prior on more identities (15 to 750) (ablation)Preface prior · Studio dataset, 2 views, 1KPSNR: 24.25 -> 25.43Improved
also optimise latent code during fine-tuning rather than model weights only (ablation)Preface fine-tuning · Studio dataset, 2 views, 1K, subjects A/B/CPSNR: 26.07 -> 26.55 (subject A)Improved
use Preface prior instead of KeypointNeRF at 2 viewsfew-shot face NVS · Studio dataset, 2 views, 1KLPIPS: 0.2713 -> 0.1905Improved
Preface (not trained on FaceScape) compared with DINER on FaceScape subject 344Preface · FaceScape, 4 viewsPSNR: 25.51 vs 26.45 (DINER)Mixed

Earlier methods this paper compares with

Named baselines from earlier accepted papers in this paper's results tables: better than 1 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