AI Paper Atlas

Research map / Detection and segmentation / Open-set and few-shot detection

Visual Relationship Detection with Language Priors

Cewu Lu, Ranjay Krishna, Michael Bernstein, Li Fei-Fei

ECCV 2016 · Detection and segmentation › Open-set and few-shot detection · approach: Visual relationship and scene graph detection · 1,148 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
add K() embedding term that pulls similar relationships together, on top of V+L (ablation)Ours - V + L · Visual Relationship Detection dataset, predicate detectionRecall@100: 35.20 -> 47.87Improved
add K() embedding term on top of V+L (ablation)Ours - V + L · Visual Relationship Detection dataset, phrase detectionRecall@100: 8.59 -> 17.03Improved
add K() embedding term on top of V+L (ablation)Ours - V + L · Visual Relationship Detection dataset, relationship detectionRecall@100: 9.18 -> 14.70Improved
use K() instead of a plain L2 regularizer on W (ablation)Ours - V + L + Reg. · Visual Relationship Detection dataset, relationship detectionRecall@100: 9.63 -> 14.70Improved
add language likelihood L to the visual module (ablation)Ours - V only · Visual Relationship Detection dataset, predicate detectionRecall@100: 7.11 -> 35.20Improved
add visual module V to the language-only model (ablation)Ours - L only · Visual Relationship Detection dataset, relationship detectionRecall@100: 0.08 -> 9.18Improved
use learned projection function f() instead of naive training-frequency function (ablation)Ours - V + naive FC · Visual Relationship Detection dataset, predicate detectionRecall@100: 28.87 -> 35.20Improved
add K() embedding term for unseen relationships (ablation)Ours - V + L · Visual Relationship Detection dataset (zero-shot), relationship detectionRecall@100: 2.66 -> 3.52Improved
add K() embedding term for unseen relationships (ablation)Ours - V + L · Visual Relationship Detection dataset (zero-shot), predicate detectionRecall@100: 6.11 -> 8.45Improved
add language likelihood L to visual module for unseen relationships (ablation)Ours - V only · Visual Relationship Detection dataset (zero-shot), phrase detectionRecall@100: 1.12 -> 2.56Improved
add K() embedding term on Visual Phrases dataset (ablation)Ours V + L only · Visual Phrases dataset (12 phrases), phrase detectionmAP: 57.3 -> 59.2Improved
add language likelihood L on Visual Phrases dataset (ablation)Ours V only · Visual Phrases dataset (12 phrases), phrase detectionmAP: 53.4 -> 57.3Improved

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