English (United Kingdom)

https://curated-unify.zendy.io/wp-json/zendy-region/v1/featured_content/oa?rat=en

https://curated-unify.zendy.io/wp-json/zendy-region/v1/highlighted_journal/

Global: Zendy Plus annual

Presents the access of premium content as premium feature

Premium Content

Presents the keyphrase highlighting as premium feature

Keyphrase Highlighting

Presents the summarisation as premium feature

Summarisation

Insights

Presents the pdf analysis as premium feature

PDF Analysis

Presents the zaia usage as premium feature

ZAIA

Global: Zendy Plus monthly

Global: Zendy Tools annual

Global: Zendy Tools monthly

Global: Zendy Free Plan

Given a large unlabeled set of images, how to efficiently and effectivelygroup them into clusters based on extracted visual representations remains achallenging problem. To address this problem, we propose a convolutional neuralnetwork (CNN) to jointly solve clustering and representation learning in aniterative manner. In the proposed method, given an input image set, we firstrandomly pick k samples and extract their features as initial cluster centroidsusing the proposed CNN with an initial model pre-trained from the ImageNetdataset. Mini-batch k-means is then performed to assign cluster labels toindividual input samples for a mini-batch of images randomly sampled from theinput image set until all images are processed. Subsequently, the proposed CNNsimultaneously updates the parameters of the proposed CNN and the centroids ofimage clusters iteratively based on stochastic gradient descent. We alsoproposed a feature drift compensation scheme to mitigate the drift error causedby feature mismatch in representation learning. Experimental resultsdemonstrate the proposed method outperforms start-of-the-art clustering schemesin terms of accuracy and storage complexity on large-scale image setscontaining millions of images.

CNN-Based Joint Clustering and Representation Learning with Feature Drift Compensation for Large-Scale Image Data