AI Paper Atlas

Research map / Image generation / Editing and style transfer

Towards Universal Image Embeddings: A Large-Scale Dataset and Challenge for Generic Image Representations

Nikolaos-Antonios Ypsilantis, Kaifeng Chen, Bingyi Cao, Mário Lipovský, Pelin Dogan-Schönberger, Grzegorz Makosa, Boris Bluntschli, Mojtaba Seyedhosseini, Ondřej Chum, André Araujo

ICCV 2023 · Image generation › Editing and style transfer · approach: Image-to-image translation · 16 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
train a single universal model on all 8 domains instead of an oracle assembly of specialistsCLIP ViT-B/16 64-D · UnED test (mean)mMP@5: 59.6 -> 55.0Hurt
finetune on UnED training set instead of off-the-shelf 64-D PCA-whitened CLIPCLIP ViT-B/16 · UnED test (mean)mMP@5: 31.0 -> 55.0Improved
use PCA-whitening to reduce off-the-shelf embedding to 64-D (ablation)CLIP and IN embeddings · UnED test (mean)mMP@5: 39.8 -> 31.0 (CLIP)Hurt
sample domains round-robin or by specialist steps instead of by dataset size (ablation)universal CLIP training · UnED test (mean)mMP@5: 52.4 -> 55.0Improved
use CLIP initialization instead of ImageNet for universal training (ablation)universal embeddings · UnED test (mean)mMP@5: 51.3 -> 55.0 (USCSS)Improved
use a joint classifier instead of separate classifiers (ablation)universal CLIP training with round-robin sampling · UnED test, SOP and InShop domainsR@1: InShop 89.4 (separate) -> 91.1 (joint)Mixed
jointly train Met with GLDv2 in a universal modeluniversal embedding · UnED test, Met domainmMP@5: 27.6 -> 2.5Hurt
finetune CLIP ViT-H/14 with stratified learning rates (projection/classifier 10^3 larger than backbone)CLIP ViT-H/14 · Universal Embedding Challenge private splitmMP@5: 62.1 -> 70.9Improved
use bigger CLIP ViT-H/14 challenge solutions on UnED instead of paper baselines in some domainschallenge solutions · UnED test (CARS196, Met, GLDv2, Rp2k)mMP@5/R@1Improved

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