AI Paper Atlas

Research map / Image generation / Image restoration

QC-StyleGAN - Quality Controllable Image Generation and Manipulation

Dat Viet Thanh Nguyen, Phong Tran The, Tan M. Dinh, Cuong Pham, Anh Tuan Tran

NeurIPS 2022 · Image generation › Image restoration · approach: Priors for restoration networks · 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
Remove distillation loss, no freezing (ablation)QC-StyleGAN · FFHQFID sharp: 3.65 -> 11.32Hurt
Remove distillation loss, with freezing (ablation)QC-StyleGAN · FFHQFID sharp: 3.65 -> 33.45Hurt
Finetune only the last 1 synthesis block (M=1) instead of 2 (ablation)QC-StyleGAN · FFHQFID sharp: 6.38Hurt
Stack P=5 conv layers in DegradBlock instead of 3 (ablation)QC-StyleGAN · FFHQFID sharp: 4.02Hurt
Reduce quality code size to Dq=8 with L=64 (ablation)QC-StyleGAN · FFHQFID sharp: 4.54Hurt
AdaIN-style DegradBlock instead of PCA-based (ablation)QC-StyleGAN · FFHQFID sharp: 4.41Hurt
Inject q by concatenating with z, w, or DualStyleGAN structure (ablation)QC-StyleGAN · FFHQcontent match of sharp/degradedHurt
Train degraded-image StyleGAN2-Ada separately vs QC-StyleGANStyleGAN2-Ada · FFHQ degraded generationFID: 4.38 -> 3.23Improved
QC-StyleGAN versus SG2-Ada on sharp generationStyleGAN2-Ada · FFHQ sharpFID: 3.48 -> 3.65No clear effect
QC-StyleGAN versus SG2-Ada on LSUN-Church sharpStyleGAN2-Ada · LSUN-Church sharpFID: 3.86 -> 3.61Improved
Use QC-StyleGAN for multiple-degradation restorationimage restoration baselines · CelebA-HQ multi-degNIQE, LPIPS: 8.97 -> 5.64 NIQE (vs MPRNet)Improved
QC-StyleGAN restoration versus specialized methods on noise LPIPSDnCNN · CelebA-HQ noiseLPIPS: 0.080 vs 0.183Hurt

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