Cardiff University | Prifysgol Caerdydd ORCA
Online Research @ Cardiff 
WelshClear Cookie - decide language by browser settings

Contextual hashing with decoupled localization for efficient video corpus moment retrieval

Kuai, Mingjin, Peng, Jin, Lv, Zheqi, Qin, You, Yang, Zhan, Zhang, Zhen, Zhou, Wei and Ji, Wei 2026. Contextual hashing with decoupled localization for efficient video corpus moment retrieval. IEEE Transactions on Image Processing 35 , pp. 11392-11404. 10.1109/tip.2026.3746991

Full text not available from this repository.

Abstract

The aim of Video Corpus Moment Retrieval (VCMR) is to locate temporal moments within a large video corpus on the basis of textual queries; however, it faces challenges in balancing retrieval efficiency and localization accuracy. Existing approaches typically rely on high-dimensional continuous embeddings for retrieval and computationally intensive coupled interaction networks for localization, thereby resulting in severe bottlenecks. To resolve this challenge, we propose a novel framework named Contextual Hashing with Decoupled Localization (CHDL). First, to address the efficiency issue, we introduce a temporal contextual hashing mechanism. This module projects high-dimensional video-text features into compact binary codes while preventing feature collapse through entropy maximization, thereby enabling ultra-fast retrieval through hardware-friendly bit-wise Hamming distance computation. Second, we propose a decoupled multi-scale moment locator, that structurally decouples cross-modal semantic interactions from downstream temporal boundary prediction. By operating exclusively on one-dimensional relevance signals derived from a token-adaptive dynamic relevance model and using pyramidal dilated convolutions, this lightweight locator achieves precise temporal localization with minimal computational overhead. Furthermore, to ensure robust feature discriminability, we design a representation alignment objective that is based on dual-granularity mutual information maximization. This objective enforces alignment at both the global query-video level and the local token-frame level, thereby facilitating high-quality cross-modal interaction. The results of extensive experiments on the TVR and DiDeMo-FIG datasets demonstrate that CHDL achieves not only superior performance in localization accuracy but also a significant reduction in retrieval latency.

Item Type: Article
Date Type: Publication
Status: Published
Schools: Schools > Computational & Mathematical Sciences
Schools > Computer Science & Informatics
Publisher: Institute of Electrical and Electronics Engineers (IEEE)
ISSN: 1057-7149
Last Modified: 09 Nov 2026 12:15
URI: https://orca.cardiff.ac.uk/id/eprint/190763

Actions (repository staff only)

Edit Item Edit Item