AI Paper Atlas

Research map / Image generation / Editing and style transfer

Learning the Latent "Look": Unsupervised Discovery of a Style-Coherent Embedding From Fashion Images

Wei-Lin Hsiao, Kristen Grauman

ICCV 2017 · Image generation › Editing and style transfer · approach: Arbitrary neural style transfer · 104 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
use polylingual LDA instead of monolingual LDA (ablation)style topic model · HipsterWars (GT attributes)avg max AP: 0.48 -> 0.53Improved
use polylingual LDA instead of monolingual LDA (ablation)style topic model · HipsterWars (GT attributes)NMI: 0.30 -> 0.31No clear effect
use ground truth attributes instead of predicted attributes (ablation)PolyLDA · DeepFashion style discoveryavg AP: 0.0647 -> 0.1762Improved
use ground truth attributes instead of predicted attributes (ablation)PolyLDA · HipsterWars style discoveryavg AP: 0.50 -> 0.53Improved
use PolyLDA topics instead of raw attribute indicator vectors (predicted attributes)Attributes · HipsterWars style discoveryavg AP: 0.28 -> 0.50Improved
use PolyLDA topics instead of StyleNet featuresStyleNet · DeepFashion style discoveryNMI: 0.0011 -> 0.0116Improved
fine-tune ResNet-50 on 148 attributes (Attr-ResNet) instead of vanilla ImageNet ResNet-50ResNet-50 · HipsterWars style discoveryavg AP: 0.30 -> 0.35Improved
fine-tune CNN on fashion metadata labels (StyleNet) instead of vanilla ResNetResNet-50 · HipsterWars style discoveryavg AP: 0.30 -> 0.39Improved
choose clustering algorithm (K-means, GMM, AP-clustering) for baseline embeddings (ablation)baseline CNN features · style discoveryaccuracyNo clear effect
vary number of topics K from 15 to 30 in style mixing (ablation)PolyLDA · style mixingAPNo clear effect
use PolyLDA instead of StyleNet for mixing styles (GT attributes)StyleNet · Style mixing, Bohemian x HipsterAP: 0.113 -> 0.394Improved
use PolyLDA with predicted attributes instead of Attributes baseline for Hipster x Goth mixingAttributes · Style mixing, Hipster x GothAP: 0.172 -> 0.424Improved

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