atlas

Research map / Image generation / Editing and style transfer

MetaStyle: Three-Way Trade-off among Speed, Flexibility, and Quality in Neural Style Transfer

AAAI 2019 · Image generation › Editing and style transfer · approach: Arbitrary neural style transfer · 32 citations (OpenAlex)

Read the paperCheck your idea against 108,967 papers

AI-extracted summary. Written by AI from the paper's title only (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 style-neutral initialization with bilevel optimization and adapt it to a style for 200 iterationsMetaStyle · Shared style comparison with Johnson et al.iterations to obtain a high-quality style transfer model: 160,000 -> 200Helped
Initialize Gatys-style optimization with the MetaStyle style-neutral representation instead of the content imageGatys-style optimization · Style transfer comparisonconvergence qualityHelped
Use bilevel meta-learning with a small number of style-specific update stepsMetaStyle · Arbitrary style transferperceptual lossHelped
Adapt the neutral representation to a new style rather than train a single-style model from scratchMetaStyle · Arbitrary style transferadaptation time: 200 iterations; 24 secondsHelped

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