AI Paper Atlas

Research map / Image generation / Editing and style transfer

Old Photo Restoration via Deep Latent Space Translation

Ziyu Wan, Bo Zhang, Dong Chen, Pan Zhang, Dong Chen, Fang Wen, Jing Liao

TPAMI 2022 · Image generation › Editing and style transfer · approach: GAN inversion for editing · 95 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 partial nonlocal block (ablation)triplet translation network · DIV2K syntheticFID: 143.62 -> 134.35Improved
add partial nonlocal block (ablation)triplet translation network · DIV2K syntheticPSNR: 23.14 -> 23.33Improved
add partial nonlocal block (ablation)triplet translation network · DIV2K syntheticLPIPS: 0.26 -> 0.25Improved
use VAE latent space instead of Pix2Pix (ablation)translation network · Real old photosWasserstein distance: 1.837 -> 1.048Improved
add two-stage training of VAEs (ablation)VAE latent translation · Real old photosWasserstein distance: 1.048 -> 0.765Improved
add latent adversarial loss (ablation)VAE latent translation · Real old photosWasserstein distance: 0.765 -> 0.581Improved
add latent adversarial loss (ablation)VAE latent translation · Real old photosBRISQUE: 23.396 -> 23.016Improved
use hierarchical injection instead of single 256x256 injection (ablation)face enhancement network · Face enhancementFID: 15.653 -> 13.175Improved
use hierarchical injection instead of single 256x256 injection (ablation)face enhancement network · Face enhancementPSNR: 24.938 -> 25.282Improved
increase single injection resolution from 16x16 to 128x128 (ablation)face enhancement network · Face enhancementFID: 24.177 -> 14.236Improved
inject only at the highest 256x256 scale rather than 128x128 (ablation)face enhancement network · Face enhancementFID: 14.236 -> 15.653Hurt
add joint training of face enhancement with translation network (ablation)face enhancement · Real old photosvisual 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