Go to page

Bibliographic Metadata

End-to-End Cross-Modality Retrieval with CCA Projections and Pairwise Ranking Loss
AuthorDorfer, Matthias ; Schlüter, Jan ; Vall, Andreu ; Korzeniowski, Filip ; Widmer, Gerhard
Published in
International Journal of Multimedia Information Retrieval, 2018, Vol. 7, Issue 2, page 117-128
PublishedSpringer London, 2018
Document typeJournal Article
Keywords (EN)Cross-modality retrieval / Canonical correlation analysis / Ranking loss / Neural network / Joint embedding space
URNurn:nbn:at:at-ubl:3-898 Persistent Identifier (URN)
 The work is publicly available
End-to-End Cross-Modality Retrieval with CCA Projections and Pairwise Ranking Loss [1.08 mb]
Abstract (English)

Cross-modality retrieval encompasses retrieval tasks where the fetched items are of a different type than the search query, e.g., retrieving pictures relevant to a given text query. The state-of-the-art approach to cross-modality retrieval relies on learning a joint embedding space of the two modalities, where items from either modality are retrieved using nearest-neighbor search. In this work, we introduce a neural network layer based on canonical correlation analysis (CCA) that learns better embedding spaces by analytically computing projections that maximize correlation. In contrast to previous approaches, the CCA layer allows us to combine existing objectives for embedding space learning, such as pairwise ranking losses, with the optimal projections of CCA. We show the effectiveness of our approach for cross-modality retrieval on three different scenarios (text-to-image, audio-sheet-music and zero-shot retrieval), surpassing both Deep CCA and a multi-view network using freely learned projections optimized by a pairwise ranking loss, especially when little training data is available (the code for all three methods is released at: https://github.com/CPJKU/cca_layer).

The PDF-Document has been downloaded 2 times.
CC-BY-License (4.0)Creative Commons Attribution 4.0 International License