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

The success of current deep saliency detection methods heavily depends on theavailability of large-scale supervision in the form of per-pixel labeling. Suchsupervision, while labor-intensive and not always possible, tends to hinder thegeneralization ability of the learned models. By contrast, traditionalhandcrafted features based unsupervised saliency detection methods, even thoughhave been surpassed by the deep supervised methods, are generallydataset-independent and could be applied in the wild. This raises a naturalquestion that "Is it possible to learn saliency maps without using labeled datawhile improving the generalization ability?". To this end, we present a novelperspective to unsupervised saliency detection through learning from multiplenoisy labeling generated by "weak" and "noisy" unsupervised handcraftedsaliency methods. Our end-to-end deep learning framework for unsupervisedsaliency detection consists of a latent saliency prediction module and a noisemodeling module that work collaboratively and are optimized jointly. Explicitnoise modeling enables us to deal with noisy saliency maps in a probabilisticway. Extensive experimental results on various benchmarking datasets show thatour model not only outperforms all the unsupervised saliency methods with alarge margin but also achieves comparable performance with the recentstate-of-the-art supervised deep saliency methods.

Deep Unsupervised Saliency Detection: A Multiple Noisy Labeling Perspective