AI Paper Atlas

Research map / Image generation / Editing and style transfer

Learning Input-agnostic Manipulation Directions in StyleGAN with Text Guidance

Yoonjeon Kim, Hyunsu Kim, Junho Kim, Yunjey Choi, Eunho Yang

ICLR 2023 · Image generation › Editing and style transfer · approach: Instruction-based diffusion editing · 0 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
Learn a dictionary using multi-channel manipulation effectsStyleGAN2-ADA text-guided manipulation · FFHQ, 10–100 channelsR_CLIP: 0.224 -> 0.242; 0.228 -> 0.249; 0.232 -> 0.251; 0.237 -> 0.254Improved
Learn the dictionary using multi-channel manipulation effectsStyleGAN2-ADA direction recovery · FFHQ, 100 channelsMSE direction reconstruction: 0.0315 -> 0.0057Improved
Learn the dictionary using multi-channel manipulation effectsStyleGAN2-ADA direction recovery · FFHQ, zero direction channelsMSE disentanglement: 0.006218 -> 0.00486Improved
Use Multi2One instead of StyleCLIP GlobalDirectionStyleGAN2-ADA text-guided manipulation · FFHQ, 57 text promptsR_CLIP: 0.2482 -> 0.2617Improved
Use Multi2One instead of StyleCLIP Local OptimizationStyleGAN2-ADA text-guided manipulation · FFHQ, 57 text promptsR_CLIP: 0.2698 -> 0.2617Hurt
Add the multi-channel CLIP reconstruction loss (λ=0.01) (ablation)Multi2One · FFHQ, 10–100 channelsR_CLIP: 0.235 -> 0.242; 0.240 -> 0.249; 0.243 -> 0.251; 0.248 -> 0.254Improved
Use supervised InterFaceGAN directions instead of unsupervised directions (ablation)Multi2One dictionary learning · FFHQ, 100 channelsR_CLIP: 0.254 -> 0.228Hurt
Use top-k sparsity levels 10, 30, and 50 rather than including 100 (ablation)Multi2One dictionary learning · FFHQ, 57 text promptsR_CLIP increase: 0.121 -> 0.141Improved
Use top-k levels 10 and 30 only rather than 10, 30, and 50 (ablation)Multi2One dictionary learning · FFHQ, 57 text promptsR_CLIP increase: 0.141 -> 0.140No clear effect
Use top-k levels 10, 30, and 50 rather than top-k 10 only (ablation)Multi2One dictionary learning · FFHQ, 57 text promptsR_CLIP increase: 0.106 -> 0.141 (λ=0.01); 0.125 -> 0.115 (λ=0.1)Mixed

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