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

Face clustering is an essential task in computer vision due to the explosionof related applications such as augmented reality or photo album management.The main challenge of this task lies in the imperfectness of similarities amongimage feature representations. Given an existing feature extraction model, itis still an unresolved problem that how can the inherent characteristics ofsimilarities of unlabelled images be leveraged to improve the clusteringperformance. Motivated by answering the question, we develop an effectiveunsupervised method, named as FaceMap, by formulating face clustering as aprocess of non-overlapping community detection, and minimizing the entropy ofinformation flows on a network of images. The entropy is denoted by the mapequation and its minimum represents the least description of paths among imagesin expectation. Inspired by observations on the ranked transition probabilitiesin the affinity graph constructed from facial images, we develop an outlierdetection strategy to adaptively adjust transition probabilities among images.Experiments with ablation studies demonstrate that FaceMap significantlyoutperforms existing methods and achieves new state-of-the-arts on threepopular large-scale datasets for face clustering, e.g., an absolute improvementof more than $10\%$ and $4\%$ comparing with prior unsupervised and supervisedmethods respectively in terms of average of Pairwise F-score. Our code ispublicly available on github.

FaceMap: Towards Unsupervised Face Clustering via Map Equation