AI Paper Atlas

Research map / Image generation / GANs

Few-Shot Image Generation with Mixup-Based Distance Learning

Chaerin Kong, Jeesoo Kim, Donghoon Han, Nojun Kwak

ECCV 2022 · Image generation › GANs · approach: Large-scale GAN latent analysis · 27 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 MixDL (latent mixup and distance learning)StyleGAN2 · Anime Face 10-shotFID: 166.0 -> 73.1Improved
add MixDLStyleGAN2 · Pokemon 10-shotFID: 257.6 -> 117.0Improved
add MixDLFastGAN · Anime Face 10-shotFID: 123.7 -> 107.6Improved
add MixDLFastGAN · Pokemon 10-shotFID: 123.5 -> 144.0Hurt
add MixDLFastGAN · Oxford Flowers 10-shotFID: 182.7 -> 180.5No clear effect
add MixDL on top of ADAStyleGAN2 + ADA · Animal-Face Dog 10-shotFID: 236.5 -> 94.1Improved
add MixDL on top of DAStyleGAN2 + DA · Face Sketches 10-shotFID: 43.1 -> 35.6Improved
replace DistanceGAN with MixDLStyleGAN2 · FFHQ-babies 10-shotFID: 105.7 -> 83.4Improved
use MixDL instead of CDC pretrained on FFHQStyleGAN2 · Obama 10-shotFID: 75.0 -> 62.7Improved
add MixDL to StyleGAN2StyleGAN2 · Landscape interpolationPPL std: 12.66 -> 4.19Improved
use 8 FC mapping layers instead of 2 in DiffAugDiffAug · Obama 100-shotFID: 46.87 -> 71.13Hurt
add discriminator feature alignment (MixDL-D) to MixDL-G (ablation)MixDL · few-shot datasetsFID/diversityMixed

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