AI Paper Atlas

Research map / Vision-language and multimodal / Human and embodied modeling

Learning Joint Visual Semantic Matching Embeddings for Language-guided Retrieval

Yanbei Chen, Loris Bazzani

ECCV 2020 · Vision-language and multimodal › Human and embodied modeling · approach: Composed and text-based image retrieval · 78 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 JVSM privileged-information losses to TIRGTIRG · Fashion-200kR@1: 15.1 -> 19.0Improved
JVSM instead of TIRG*TIRG* · UT-Zap50KR@10: 25.4 -> 37.1Improved
JVSM instead of TIRG*TIRG* · Fashion-IQ DressR@10: 7.3 -> 10.7Improved
add visual semantic embedding loss Lvse and Lim to language-guided retrieval baseline (Lim only) (ablation)TIRG* · Fashion-200k language-guided image retrievalR@10: 41.9 -> 44.0Improved
add compositional text matching loss Ltm (ablation)JVSM (Lvse + Lim) · Fashion-200k language-guided image retrievalR@1: 15.6 -> 19.0Improved
add compositional text matching loss Ltm (ablation)JVSM (Lvse + Lim) · Fashion-200k language-guided text retrievalR@1: 34.2 -> 50.4Improved
add compositional losses to VSE-only model (ablation)VSE (Lvse) · Fashion-200k text-to-image retrievalR@1: 22.7 -> 23.4Improved
add Lim only to VSE model for text-to-image retrieval (ablation)VSE (Lvse) · Fashion-200k text-to-image retrievalR@1: 22.7 -> 21.3Hurt
use semi-hard negative mining instead of hardest negative mining (ablation)VSE training · Fashion-200kconvergenceImproved
add two-stage training (VSE first, then full loss) (ablation)JVSM · Fashion-200kloss minimumImproved
use soft semantic matching instead of semi-hard matching when privileged info is partial (ablation)JVSM · Fashion-IQperformanceImproved
use soft semantic matching instead of semi-hard matching when fully annotated (ablation)JVSM · fully annotated datasetsperformanceNo clear effect

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