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 Tools annual

Global: Zendy Free Plan

Global: Zendy Tools monthly

Global: Zendy Plus monthly

Scene recognition based on deep-learning has made significant progress, butthere are still limitations in its performance due to challenges posed byinter-class similarities and intra-class dissimilarities. Furthermore, priorresearch has primarily focused on improving classification accuracy, yet it hasgiven less attention to achieving interpretable, precise scene classification.Therefore, we are motivated to propose EnTri, an ensemble scene recognitionframework that employs ensemble learning using a hierarchy of visual features.EnTri represents features at three distinct levels of detail: pixel-level,semantic segmentation-level, and object class and frequency level. Byincorporating distinct feature encoding schemes of differing complexity andleveraging ensemble strategies, our approach aims to improve classificationaccuracy while enhancing transparency and interpretability via visual andtextual explanations. To achieve interpretability, we devised an extensionalgorithm that generates both visual and textual explanations highlightingvarious properties of a given scene that contribute to the final prediction ofits category. This includes information about objects, statistics, spatiallayout, and textural details. Through experiments on benchmark sceneclassification datasets, EnTri has demonstrated superiority in terms ofrecognition accuracy, achieving competitive performance compared tostate-of-the-art approaches, with an accuracy of 87.69%, 75.56%, and 99.17% onthe MIT67, SUN397, and UIUC8 datasets, respectively.

EnTri: Ensemble Learning with Tri-level Representations for Explainable Scene Recognition