Central Similarity Quantization for Efficient Image and Video Retrieval

By Li Yuan, Xiaopeng Zhang, Francis E. H. Tay, Zequn Jie, Wei David Liu, Jiashi Feng. Computer Vision and Pattern Recognition, 2019

374 citations recorded by the source corpus.

Existing data-dependent hashing methods usually learn hash functions from pairwise or triplet data relationships, which only capture the data similarity locally, and often suffer from low learning efficiency and low collision rate. In this work, we propose a new \emph{global} similarity metric, termed as \emph{central similarity}, with which the hash codes of similar data pairs are encouraged to approach a common center and those for dissimilar pairs to converge to different centers, to improve hash learning efficiency and retrieval accuracy. We principally formulate the computation of the proposed central similarity metric by introducing a new concept, i.e., \emph{hash center} that refers to a set of data points scattered in the Hamming space with a sufficient mutual distance between each other. We then provide an efficient method to construct well separated hash centers by leveraging the Hadamard matrix and Bernoulli distributions. Finally, we propose the Central Similarity Quantization (CSQ) that optimizes the central similarity between data points w.r.t.\ their hash centers instead of optimizing the local similarity. CSQ is generic and applicable to both image and video hashing scenarios. Extensive experiments on large-scale image and video retrieval tasks demonstrate that CSQ can generate cohesive hash codes for similar data pairs and dispersed hash codes for dissimilar pairs, achieving a noticeable boost in retrieval performance, i.e. 3\%-20\% in mAP over the previous state-of-the-arts.

Subfields: Computer Vision and Pattern Recognition

View Paper: https://www.semanticscholar.org/paper/b0bc3af8369ad22ed7a22af247d6c9b86e888170

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Computer Vision and Pattern Recognition

Citations in the corpus

Toward Accurate Person-level Action Recognition in Videos of Crowed Scenes (2020, methodology): “Since RCNN [8] has been proposed, the two-stage detection methods have been widely adopted or modified [1, 14, 16, 23, 27, 28, 31, 32, 36].”
A review of convolutional neural network architectures and their optimizations (2022, neutral): “In 2020, Yuan et al. put forward a novel central similarity measure, i.e., Central Similarity Quantization (CSQ) (Yuan et al.”
A Simple Baseline for Pose Tracking in Videos of Crowed Scenes (2020, methodology): “Since RCNN [5] has been proposed, the two-stage detection methods have been widely adopted or modified [1, 9, 11, 17, 21, 22, 27, 28, 30].”
Exponential Moving Average Normalization for Self-supervised and Semi-supervised Learning (2021, methodology): “We used train as the retrieval database and val as queries, and followed [49] to use the top 1000 retrievals for the evaluation of mean averaged precision (mAP) and recall.”