HyperAI
StartseiteNeuigkeitenNeueste ForschungsarbeitenTutorialsDatensätzeWikiSOTALLM-ModelleGPU-RanglisteVeranstaltungen
Suche
Über
HyperAI

Zero Shot Composed Image Retrieval Zs Cir On 11

Metriken

A-R@1

Ergebnisse

Leistungsergebnisse verschiedener Modelle zu diesem Benchmark

Modellname
A-R@1
Paper TitleRepository
LinCIR (CLIP L/14)12.2Language-only Efficient Training of Zero-shot Composed Image Retrieval
OSrCIR (CLIP B/32)17.4Reason-before-Retrieve: One-Stage Reflective Chain-of-Thoughts for Training-Free Zero-Shot Composed Image Retrieval
SEARLE (CLIP B/32)14.4Zero-Shot Composed Image Retrieval with Textual Inversion
LinCIR (CLIP G/14)13.7Language-only Efficient Training of Zero-shot Composed Image Retrieval
SEARLE (CLIP L/14)14.4Zero-Shot Composed Image Retrieval with Textual Inversion
OSrCIR (CLIP L/14)17.9Reason-before-Retrieve: One-Stage Reflective Chain-of-Thoughts for Training-Free Zero-Shot Composed Image Retrieval
OSrCIR (CLIP G/14)19.6Reason-before-Retrieve: One-Stage Reflective Chain-of-Thoughts for Training-Free Zero-Shot Composed Image Retrieval
CIReVL (CLIP G/14)-Vision-by-Language for Training-Free Compositional Image Retrieval
Context-I2W (CLIP L/14)12.7Context-I2W: Mapping Images to Context-dependent Words for Accurate Zero-Shot Composed Image Retrieval
CIReVL (CLIP L/14)-Vision-by-Language for Training-Free Compositional Image Retrieval
CIReVL (CLIP B/32)-Vision-by-Language for Training-Free Compositional Image Retrieval
0 of 11 row(s) selected.