SOTAVerified

Zero-Shot Composed Image Retrieval (ZS-CIR)

Given a query composed of a reference image and a relative caption, Composed Image Retrieval (CIR) aims to retrieve target images that are visually similar to the reference one but incorporate the changes specified in the relative caption. The bi-modality of the query provides users with more precise control over the characteristics of the desired image, as some features are more easily described with language, while others can be better expressed visually.

Zero-Shot Composed Image Retrieval (ZS-CIR) is a subtask of CIR that aims to design an approach that manages to combine the reference image and the relative caption without the need for supervised learning.

Papers

Showing 2130 of 36 papers

TitleStatusHype
iSEARLE: Improving Textual Inversion for Zero-Shot Composed Image RetrievalCode2
Improving Composed Image Retrieval via Contrastive Learning with Scaling Positives and NegativesCode1
MagicLens: Self-Supervised Image Retrieval with Open-Ended InstructionsCode3
Knowledge-Enhanced Dual-stream Zero-shot Composed Image RetrievalCode1
Training-free Zero-shot Composed Image Retrieval with Local Concept Reranking0
Language-only Efficient Training of Zero-shot Composed Image RetrievalCode1
Pretrain like Your Inference: Masked Tuning Improves Zero-Shot Composed Image RetrievalCode0
Vision-by-Language for Training-Free Compositional Image RetrievalCode1
Context-I2W: Mapping Images to Context-dependent Words for Accurate Zero-Shot Composed Image RetrievalCode1
CoVR-2: Automatic Data Construction for Composed Video RetrievalCode1
Show:102550
← PrevPage 3 of 4Next →

No leaderboard results yet.