Skip to content

Prototype-driven scale-aware modeling for multimodal hashing

Aug 2026 · Multimedia Systems · Vol 32 · 0 citations · 57 references

TL;DR

A prototype-driven scale-aware modeling multi-modal hashing framework that jointly models cross-modal interaction and hierarchical prototype learning in an end-to-end manner is proposed, enabling the learned hash codes to better preserve fine-grained and overlapping semantic structures commonly observed in real-world multimodal data.

View source

Similar papers

Oct 2026

SSMH: Semantic Self-refinement Multi-modal Hashing for Multi-modal Retrieval

Multi-modal hashing converts high-dimensional features into compact binary codes while preserving semantic similarity, thereby achieving high performance with lower storage and computational costs. However, existing methods generally overly depend on discrete supervisory information, ignoring the ambiguity of labels an...

Chao-Qun Zheng, Rui-Fan Zuo, Feng-Ling Li et al. · 0 citations
Open access Oct 2026

Triple collaborative asymmetric deep hashing for multi-label image retrieval

Due to its efficient storage and fast retrieval capabilities, hashing methods have garnered significant attention in image retrieval applications. However, most existing methods rely on semantic similarity matrices to indicate whether two images are similar or not, which neglects the guiding role of semantic signals...

Wei-Gang Wang, Zi-Yuan Cui · 0 citations
Sep 2026

Do More With Less: Capacity-Aware Selective Hashing for Continual Cross-Modal Retrieval.

Capacity-Aware Selective Hashing (CASH) is proposed, which significantly improves Hamming-space utilization through bit-level selective allocation under a fixed capacity budget, enabling stable continual learning while preserving long-term code compatibility.

Wen-Hao Liu, Tian-Shi Wang, Feng-Ling Li et al. · 0 citations
Open access 2026

MDH: Mask-Induced Decoupled Hashing for Fine-Grained Image Retrieval

This work designs an efficient foreground-background decoupling mechanism that separates foreground and background regions at a coarse level through mask prediction, where foreground features focus on the most discriminative core regions of the category, while background features supplement rich contextual information.

Yang-Tao Wang, Hong-Xing You, Yan-Zhao Xie et al. · 0 citations

We use cookies to run the site and, with your consent, for analytics and to show ads. See our Cookie Policy.