Papers › TMCIR: Token Merge Benefits Composed Image Retrieval

TMCIR: Token Merge Benefits Composed Image Retrieval

15 Apr 2025arXiv:2504.10995archive 2025-07-28

Chaoyang Wang, Zeyu Zhang, Long Teng, Zijun Li, Shichao Kan

Composed Image Retrieval (CIR) retrieves target images using a multi-modal query that combines a reference image with text describing desired modifications. The primary challenge is effectively fusing this visual and textual information. Current cross-modal feature fusion approaches for CIR exhibit an inherent bias in intention interpretation. These methods tend to disproportionately emphasize either the reference image features (visual-dominant fusion) or the textual modification intent (text-dominant fusion through image-to-text conversion). Such an imbalanced representation often fails to accurately capture and reflect the actual search intent of the user in the retrieval results. To address this challenge, we propose TMCIR, a novel framework that advances composed image retrieval through two key innovations: 1) Intent-Aware Cross-Modal Alignment. We first fine-tune CLIP encoders contrastively using intent-reflecting pseudo-target images, synthesized from reference images and textual descriptions via a diffusion model. This step enhances the encoder ability of text to capture nuanced intents in textual descriptions. 2) Adaptive Token Fusion. We further fine-tune all encoders contrastively by comparing adaptive token-fusion features with the target image. This mechanism dynamically balances visual and textual representations within the contrastive learning pipeline, optimizing the composed feature for retrieval. Extensive experiments on Fashion-IQ and CIRR datasets demonstrate that TMCIR significantly outperforms state-of-the-art methods, particularly in capturing nuanced user intent.

PaperPDF

Code

No code repository is listed for this paper in the archive or in Syntology's graph.

Code Syntology ran Syntology

Not run by Syntology. Nothing on this page verifies that the listed code works.

Tasks

Contrastive LearningImage RetrievalImage to textRetrievalcross-modal alignment

Results from the paper archive 2025-07-28

TaskDatasetModelMetricValueRank at snapshotLeaderboardReport
Image Retrieval CIRR TMCIR (Recall@5+Recall_subset@1)/2 83.46 #1 of 17 Archive leaderboard report
Image Retrieval CIRR TMCIR Recall@10 91.06 #1 of 17 Archive leaderboard report
Image Retrieval Fashion IQ TMCIR (Recall@10+Recall@50)/2 66.56 #2 of 22 Archive leaderboard report
Image Retrieval Fashion IQ TMCIR Recall@10 56.57 #2 of 22 Archive leaderboard report

Ranks are positions in the archive's leaderboards as they stood at the 2025-07-28 snapshot. Results published since then are not among these rows, so a rank here is not a current standing.

Methods

CLIPContrastive LearningDiffusion

Report a problem or propose a change · a person checks every report against the paper or source before anything changes; decisions are listed on /corrections