AI Paper Atlas

Research map / 3D vision / Face and shape

An End-to-End Depth-Based Pipeline for Selfie Image Rectification

Ahmed Alhawwary, Janne Mustaniemi, Phong Nguyen-Ha, Janne Heikkilä

TPAMI 2025 · 3D vision › Face and shape · approach: Face reflectance and relightable rendering · 1 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 end-to-end training of the depth network through the differentiable renderer (ablation)depth-based rectification pipeline · Synthetic multiview datasetPSNR: 18.813 -> 19.617Improved
add end-to-end training of the depth network through the differentiable renderer (ablation)depth-based rectification pipeline · Synthetic multiview datasetSSIM: 0.7997 -> 0.8324Improved
add the horizontal translation (HzT) module (ablation)pipeline with E2E · Synthetic multiview datasetPSNR: 19.617 -> 19.894Improved
add the horizontal translation (HzT) module (ablation)pipeline with E2E · Synthetic multiview datasetSSIM: 0.8324 -> 0.8332No clear effect
use depth-based reprojection pipeline instead of LPUP flow-based pipelineLPUP · Synthetic Unreal Engine test setPSNR: 17.80 -> 19.89Improved
use depth-based reprojection pipeline instead of LPUP flow-based pipelineLPUP · Synthetic Unreal Engine test setSSIM: 0.6051 -> 0.8332Improved
use depth-based pipeline instead of Fried et al.Fried · CMDPLPIPS: 0.2128 -> 0.1919Improved
use depth-based pipeline instead of Disco 3D GAN inversionDisco · CMDPlandmark error: 0.00448 -> 0.00453No clear effect
use depth-based pipeline instead of Disco 3D GAN inversionDisco · CMDPLPIPS: 0.1934 -> 0.1919Improved
use depth-based pipeline instead of DFNDFN · CMDP (smaller subset)landmark error: 0.00613 -> 0.00499Improved
use depth-based pipeline instead of Disco inversion for speedDisco · HD image runtimeruntime: 130 s -> 0.4 sImproved
finetune pretrained depth model on synthetic facial data (ablation)pretrained monocular depth regressor · synthetic datasetdepth map qualityImproved

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