AI Paper Atlas

Research map / 3D vision / 3D generation

Specular-to-Diffuse Translation for Multi-View Reconstruction

Shihao Wu, Hui Huang, Tiziano Portenier, Matan Sela, Daniel Cohen-Or, Ron Kimmel, Matthias Zwicker

ECCV 2018 · 3D vision › 3D generation · approach: Diffusion-based novel view synthesis · 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
use S2Dnet translation instead of reconstructing directly from glossy imagesmulti-view surface reconstruction · Synthetic scenes (average)reconstruction error (% bbox diagonal): 0.90 -> 0.64Improved
use multi-view S2Dnet instead of single-view cycleGANspecular-to-diffuse translation · Synthetic scenes (average)reconstruction error (% bbox diagonal): 0.80 -> 0.64Improved
use cycleGAN translation instead of reconstructing from glossy imagesmulti-view surface reconstruction · Synthetic scenes (average)reconstruction error (% bbox diagonal): 0.90 -> 0.80Improved
use cycleGAN translation instead of glossy input on scene 1multi-view surface reconstruction · Synthetic scene 1reconstruction error (% bbox diagonal): 0.67 -> 1.18Hurt
use S2Dnet instead of cycleGAN for image translationspecular-to-diffuse translation · Synthetic test setimage MSE: 69.15 -> 57.78Improved
use supervised pix2pix instead of unsupervised S2Dnetspecular-to-diffuse translation · Synthetic test setimage MSE: 57.78 -> 56.20Improved
add perceptual correspondence loss (lambda_corr = 5 instead of 0) (ablation)S2Dnet · Synthetic sequence (qualitative)inter-view consistencyImproved
concatenate views along spatial axis instead of stacking pixel-wise (ablation)S2Dnet · Synthetic data (qualitative)translation sharpness/consistencyImproved
fine-tune or train on small real-world datasetS2Dnet · Real-world objects not seen in trainingtranslation qualityHurt
use U-Net, ResNet or RNN-block generator (ablation)S2Dnet · Synthetic datatranslation qualityNo clear effect
use paired (ConditionalGAN) instead of unpaired (cycleGAN) training (ablation)S2Dnet · Synthetic datatranslation qualityNo clear effect

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