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

Predicting attention regions of interest is an important yet challenging taskfor self-driving systems. Existing methodologies rely on large-scale labeledtraffic datasets that are labor-intensive to obtain. Besides, the huge domaingap between natural scenes and traffic scenes in current datasets also limitsthe potential for model training. To address these challenges, we are the firstto introduce an unsupervised way to predict self-driving attention byuncertainty modeling and driving knowledge integration. Our approach'sUncertainty Mining Branch (UMB) discovers commonalities and differences frommultiple generated pseudo-labels achieved from models pre-trained on naturalscenes by actively measuring the uncertainty. Meanwhile, our KnowledgeEmbedding Block (KEB) bridges the domain gap by incorporating driving knowledgeto adaptively refine the generated pseudo-labels. Quantitative and qualitativeresults with equivalent or even more impressive performance compared tofully-supervised state-of-the-art approaches across all three public datasetsdemonstrate the effectiveness of the proposed method and the potential of thisdirection. The code will be made publicly available.

Unsupervised Self-Driving Attention Prediction via Uncertainty Mining and Knowledge Embedding