Papers › Conditioned and Composed Image Retrieval Combining and Partially Fine-Tuning...

Conditioned and Composed Image Retrieval Combining and Partially Fine-Tuning CLIP-Based Features

19 Jun 2022CVPRW 2022 6archive 2025-07-28

Alberto Baldrati, Marco Bertini, Tiberio Uricchio, Alberto del Bimbo

In this paper, we present an approach for conditioned and composed image retrieval based on CLIP features. In this extension of content-based image retrieval (CBIR), an image is combined with a text that provides information regarding user intentions and is relevant for application domains like e-commerce. The proposed method is based on an initial training stage where a simple combination of visual and textual features is used, to fine-tune the CLIP text encoder. Then in a second training stage, we learn a more complex combiner network that merges visual and textual features. Contrastive learning is used in both stages. The proposed approach obtains state-of-the-art performance for conditioned CBIR on the FashionIQ dataset and for composed CBIR on the more recent CIRR dataset.

PaperPDFCode

Code

abaldrati/clip4cirdemo officialpytorch report

Repository list and official/mentioned flags are the archive's, frozen 2025-07-28. Reachability, where shown, is from one Syntology probe window (2026-09-16 to 2026-09-18); repositories not probed show nothing. GitHub stars are not tracked.

Code Syntology ran Syntology

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

Tasks

Composed Image Retrieval (CoIR)Content-Based Image RetrievalContrastive LearningImage RetrievalRetrieval

Results from the paper archive 2025-07-28

TaskDatasetModelMetricValueRank at snapshotLeaderboardReport
Image Retrieval CIRR CLIP4Cir (v2) (Recall@5+Recall_subset@1)/2 69.09 #14 of 17 Archive leaderboard report
Image Retrieval Fashion IQ CLIP4Cir (v2) (Recall@10+Recall@50)/2 50.03 #14 of 22 Archive leaderboard report
Image Retrieval LaSCo CLIP4CIR Recall@1 (%) 4.01 #3 of 3 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 Learning

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