AI Paper Atlas

Research map / Image generation / GANs

Physics-Based Generative Adversarial Models for Image Restoration and Beyond

Jinshan Pan, Jiangxin Dong, Yang Liu, Jiawei Zhang, Jimmy Ren, Jinhui Tang, Yu‐Wing Tai, Ming–Hsuan Yang

TPAMI 2020 · Image generation › GANs · approach: GAN-based image manipulation · 201 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 the physics-model constraint (second discriminator and re-degradation losses) to the GAN (ablation)BaseGAN · Image dehazing ablationPSNR: 29.90 -> 32.05Improved
add the physics-model constraint to the GAN (ablation)BaseGAN · Image dehazing ablationSSIM: 0.9541 -> 0.9722Improved
add the second ablated loss term (ablation)Ours · Image dehazing ablationPSNR: 21.24 -> 32.05Improved
add the first ablated loss term (ablation)Ours · Image dehazing ablationPSNR: 31.93 -> 32.05Improved
add the third ablated loss term (ablation)Ours · Image dehazing ablationPSNR: 31.71 -> 32.05Improved
add both of the paired loss terms (ablation)Ours · Image dehazing ablationPSNR: 30.53 -> 32.05Improved
use physics-model GAN instead of PCycleGAN with same losses (ablation)PCycleGAN · Image dehazing ablationPSNR: 27.46 -> 32.05Improved
use physics model rather than cycle consistencypix2pix · Dehazing syntheticSSIM: 0.8225 -> 0.8657Improved
use physics model rather than feed-forward CNNCNN · Text image deblurringPSNR: 26.53 -> 28.80Improved
raise lambda from 10 to 50 (ablation)Ours · Face deblurringPSNR: 22.92 -> 24.57Improved
raise lambda from 50 to 100 (ablation)Ours · Face deblurringPSNR: 24.57 -> 24.55No clear effect
increase ResBlocks from 3 to 35 (ablation)Ours · Face deblurringPSNR: 20.45 -> 21.22No 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