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

This paper focuses on visual semantic navigation, the task of producingactions for an active agent to navigate to a specified target object categoryin an unknown environment. To complete this task, the algorithm shouldsimultaneously locate and navigate to an instance of the category. Incomparison to the traditional point goal navigation, this task requires theagent to have a stronger contextual prior to indoor environments. We introduceSSCNav, an algorithm that explicitly models scene priors using aconfidence-aware semantic scene completion module to complete the scene andguide the agent's navigation planning. Given a partial observation of theenvironment, SSCNav first infers a complete scene representation with semanticlabels for the unobserved scene together with a confidence map associated withits own prediction. Then, a policy network infers the action from the scenecompletion result and confidence map. Our experiments demonstrate that theproposed scene completion module improves the efficiency of the downstreamnavigation policies. Video, code, and data: https://sscnav.cs.columbia.edu/

SSCNav: Confidence-Aware Semantic Scene Completion for Visual Semantic Navigation