UCL Discovery
UCL home » Library Services » Electronic resources » UCL Discovery

Voronoi-Based Compact Image Descriptors: Efficient Region-of-Interest Retrieval With VLAD and Deep-Learning-Based Descriptors

Chadha, A; Andreopoulos, Y; (2017) Voronoi-Based Compact Image Descriptors: Efficient Region-of-Interest Retrieval With VLAD and Deep-Learning-Based Descriptors. IEEE Transactions on Multimedia , 19 (7) pp. 1596-1608. 10.1109/TMM.2017.2673415. Green open access

[img]
Preview
Text
bare_jrnl_28.pdf - Accepted version

Download (2MB) | Preview

Abstract

We investigate the problem of image retrieval based on visual queries when the latter comprise arbitrary regionsof- interest (ROI) rather than entire images. Our proposal is a compact image descriptor that combines the state-of-the-art in content-based descriptor extraction with a multi-level, Voronoibased spatial partitioning of each dataset image. The proposed multi-level Voronoi-based encoding uses a spatial hierarchical K-means over interest-point locations, and computes a contentbased descriptor over each cell. In order to reduce the matching complexity with minimal or no sacrifice in retrieval performance: (i) we utilize the tree structure of the spatial hierarchical Kmeans to perform a top-to-bottom pruning for local similarity maxima; (ii) we propose a new image similarity score that combines relevant information from all partition levels into a single measure for similarity; (iii) we combine our proposal with a novel and efficient approach for optimal bit allocation within quantized descriptor representations. By deriving both a Voronoi-based VLAD descriptor (termed as Fast-VVLAD) and a Voronoi-based deep convolutional neural network (CNN) descriptor (termed as Fast-VDCNN), we demonstrate that our Voronoi-based framework is agnostic to the descriptor basis, and can easily be slotted into existing frameworks. Via a range of ROI queries in two standard datasets, it is shown that the Voronoibased descriptors achieve comparable or higher mean Average Precision against conventional grid-based spatial search, while offering more than two-fold reduction in complexity. Finally, beyond ROI queries, we show that Voronoi partitioning improves the geometric invariance of compact CNN descriptors, thereby resulting in competitive performance to the current state-of-theart on whole image retrieval.

Type: Article
Title: Voronoi-Based Compact Image Descriptors: Efficient Region-of-Interest Retrieval With VLAD and Deep-Learning-Based Descriptors
Open access status: An open access version is available from UCL Discovery
DOI: 10.1109/TMM.2017.2673415
Publisher version: http://doi.org/10.1109/TMM.2017.2673415
Language: English
Additional information: © 2017 IEEE. This version is the author accepted manuscript. For information on re-use, please refer to the publisher’s terms and conditions.
Keywords: Visualization, Complexity theory, Image coding, Quantization (signal), Image retrieval, Proposals, Encoding
UCL classification: UCL > Provost and Vice Provost Offices
UCL > Provost and Vice Provost Offices > UCL BEAMS
UCL > Provost and Vice Provost Offices > UCL BEAMS > Faculty of Engineering Science
UCL > Provost and Vice Provost Offices > UCL BEAMS > Faculty of Engineering Science > Dept of Electronic and Electrical Eng
URI: https://discovery.ucl.ac.uk/id/eprint/1546433
Downloads since deposit
103Downloads
Download activity - last month
Download activity - last 12 months
Downloads by country - last 12 months

Archive Staff Only

View Item View Item