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

Active recognition enables robots to intelligently explore novelobservations, thereby acquiring more information while circumventing undesiredviewing conditions. Recent approaches favor learning policies from simulated orcollected data, wherein appropriate actions are more frequently selected whenthe recognition is accurate. However, most recognition modules are developedunder the closed-world assumption, which makes them ill-equipped to handleunexpected inputs, such as the absence of the target object in the currentobservation. To address this issue, we propose treating active recognition as asequential evidence-gathering process, providing by-step uncertaintyquantification and reliable prediction under the evidence combination theory.Additionally, the reward function developed in this paper effectivelycharacterizes the merit of actions when operating in open-world environments.To evaluate the performance, we collect a dataset from an indoor simulator,encompassing various recognition challenges such as distance, occlusion levels,and visibility. Through a series of experiments on recognition and robustnessanalysis, we demonstrate the necessity of introducing uncertainties to activerecognition and the superior performance of the proposed method.

Evidential Active Recognition: Intelligent and Prudent Open-World Embodied Perception