AI Paper Atlas

Research map / Label-efficient and robust learning / Recognition and long-tail

The Missing Link: Finding Label Relations across Datasets

Jasper Uijlings, Thomas Mensink, Vittorio Ferrari

ECCV 2022 · Label-efficient and robust learning › Recognition and long-tail · approach: Partial and multi-label learning · 7 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 visual embeddings instead of pixel probabilities (ablation)visual relation discovery with set theory · ADE20k-COCO relation typeaccuracy: 47 -> 51Improved
use score asymmetry instead of set theory for relation type (ablation)Visual Embeddings · ADE20k-COCO relation typeaccuracy: 51 -> 61Improved
use score asymmetry instead of set theory for relation type (ablation)Pixel Predictions · ADE20k-COCO relation typeaccuracy: 47 -> 60Improved
combine WordNet taxonomy with visual embeddings asymmetry (ablation)Visual Embeddings asymmetry · average over dataset pairsaccuracy: 56 -> 57Improved
combine WordNet taxonomy with visual embeddings asymmetry (ablation)Visual Embeddings asymmetry · BDD-COCO relation typeaccuracy: 51 -> 53Improved
use vision instead of language onlyVisual Embeddings asymmetry vs WordNet taxonomy · average over dataset pairsaccuracy: 46 -> 56Improved
use WordNet instead of Word2Veclanguage-only relation discovery · average over dataset pairsaccuracy: 38 -> 46Improved
use pixel predictions with asymmetry vs set theory on ADE20k-BDD (ablation)Pixel Predictions · ADE20k-BDD relation typeaccuracy: 56 -> 54No clear effect
add stronger label link to predict transfer gainCOCO to ADE20k fine-tuning on 1000 images · ADE20k val per-class IoUtransfer gainImproved

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