AI Paper Atlas

Research map / Label-efficient and robust learning / Zero-/few-shot and meta-learning

Coloring With Limited Data: Few-Shot Colorization via Memory Augmented Networks

CVPR 2019 · Label-efficient and robust learning › Zero-/few-shot and meta-learning · approach: Fine-grained and cross-domain few-shot baselines · 147 citations (OpenAlex)

Check 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
Add memory networks to the colorization network (ablation)MemoPainter · Oxford102 Flower, Monster, Yumi, Superheroes, and Pokemon datasetsColorization qualityHelped
Add memory networks to the colorization network (ablation)MemoPainter · Pokemon one-shot colorizationColor identity preservationHelped
Use unsupervised threshold triplet loss instead of supervised class labelsMemoPainter memory networks · 5-way and 15-way few-shot classificationClassification accuracy: 87.50% vs 91.66%; 87.50% vs 87.50%; 69.44% vs 72.22%; 70.83% vs 75.00%Mixed
Represent stored colors with ten dominant RGB values rather than color distributionsMemoPainter · One-shot colorizationColorization performanceHelped
Use threshold triplet loss to train memory retrieval without class labelsMemoPainter memory networks · Datasets with consistent class colorsRelevance of retrieved color featuresHelped
Apply threshold triplet loss to data with arbitrary class colorsMemoPainter memory networks · Datasets where colors are arbitraryColorization performanceHurt

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