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 Tools annual

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 Tools monthly

Global: Zendy Plus monthly

Presents the access of premium content as premium feature

Premium Content

Global: Zendy Plus annual

Global: Zendy Free Plan

Discriminative correlation filters (DCF) and siamese networks have achievedpromising performance on visual tracking tasks thanks to their superiorcomputational efficiency and reliable similarity metric learning, respectively.However, how to effectively take advantages of powerful deep networks, whilemaintaining the real-time response of DCF, remains a challenging problem.Embedding the cross-correlation operator as a separate layer into siamesenetworks is a popular choice to enhance the tracking accuracy. Being a keycomponent of such a network, the correlation layer is updated online togetherwith other parts of the network. Yet, when facing serious disturbance, fusedtrackers may still drift away from the target completely due to accumulatederrors. To address these issues, we propose a coarse-to-fine trackingframework, which roughly infers the target state via an online-updating DCFmodule first and subsequently, finely locates the target through anoffline-training asymmetric siamese network (ASN). Benefitting from theguidance of DCF and the learned channel weights obtained through exploiting thegiven ground-truth template, ASN refines feature representation and implementsprecise target localization. Systematic experiments on five popular trackingdatasets demonstrate that the proposed DCF-ASN achieves the state-of-the-artperformance while exhibiting good tracking efficiency.

DCF-ASN: Coarse-to-fine Real-time Visual Tracking via Discriminative Correlation Filter and Attentional Siamese Network