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 has attracted rising research interest recently to takeadvantage of massive amounts of face images on the web. State-of-the-artperformance has been achieved by Graph Convolutional Networks (GCN) due totheir powerful representation capacity. However, existing GCN-based methodsbuild face graphs mainly according to kNN relations in the feature space, whichmay lead to a lot of noise edges connecting two faces of different classes. Theface features will be polluted when messages pass along these noise edges, thusdegrading the performance of GCNs. In this paper, a novel algorithm namedAda-NETS is proposed to cluster faces by constructing clean graphs for GCNs. InAda-NETS, each face is transformed to a new structure space, obtaining robustfeatures by considering face features of the neighbour images. Then, anadaptive neighbour discovery strategy is proposed to determine a proper numberof edges connecting to each face image. It significantly reduces the noiseedges while maintaining the good ones to build a graph with clean yet richedges for GCNs to cluster faces. Experiments on multiple public clusteringdatasets show that Ada-NETS significantly outperforms current state-of-the-artmethods, proving its superiority and generalization. Code is available athttps://github.com/damo-cv/Ada-NETS.

Ada-NETS: Face Clustering via Adaptive Neighbour Discovery in the Structure Space