AI Paper Atlas

Research map / Vision-language and multimodal / Cross-modal learning

Object-Aware Query Perturbation for Cross-Modal Image-Text Retrieval

Naoya Sogi, Takashi Shibata, Makoto Terao

ECCV 2024 · Vision-language and multimodal › Cross-modal learning · approach: Probabilistic embeddings for cross-modal retrieval · 6 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 object-aware query perturbation to BLIP2 cross-attentionBLIP2 · Flickr-30K T2I small objectsR@1: 81.33 -> 84.00Improved
add Q-Perturbation to BLIP2BLIP2 · Flickr-30K T2I overallR@1: 89.76 -> 89.86No clear effect
add Q-Perturbation to BLIP2BLIP2 · COCO 5K T2I overallR@1: 68.25 -> 68.35No clear effect
add Q-Perturbation to COCACOCA · Flickr-30K T2IR@1: 77.64 -> 77.68No clear effect
add Q-Perturbation to InternVL-GInternVL-G · Flickr-30K I2TR@1: 94.90 -> 95.40Improved
use automatically detected boxes (CO-DETR/CO-DINO) instead of annotated boxes (ablation)Q-Perturbation · Flickr-30K T2IR@1: 89.82 -> 89.86No clear effect
enhance queries in K-subspace instead of projecting onto its orthogonal complement (ablation)Q-Perturbation · Flickr-30K T2I small objectsR@1 (~10%): 77.33 -> 84.00Improved
perturb queries with K-subspace instead of adding separate object-embedding similarity scores (OEs) (ablation)BLIP2 · Flickr-30K T2IR@1: 78.48 -> 89.82Improved
vary scale, weight function and subspace dimension hyperparameters (ablation)Q-Perturbation · Flickr-30K T2ImR@1: 89.44 to 90.00No clear effect
add Q-PerturbationBLIP2 inference · visual feature extraction time on GTX 1080Timilliseconds: 321.0 -> 413.0Hurt

Earlier methods this paper compares with

Named baselines from earlier accepted papers in this paper's results tables: better than 2 of 2 on most shared results.

Matched by method name in results tables read by AI. Authors' own tables tend to favour the newer method, and a comparison can be missed when a method is renamed.

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