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

Human-Object Interaction (HOI) detection devotes to learn how humans interactwith surrounding objects. Latest end-to-end HOI detectors are short of relationreasoning, which leads to inability to learn HOI-specific interactive semanticsfor predictions. In this paper, we therefore propose novel relation reasoningfor HOI detection. We first present a progressive Relation-aware Frame, whichbrings a new structure and parameter sharing pattern for interaction inference.Upon the frame, an Interaction Intensifier Module and a Correlation ParsingModule are carefully designed, where: a) interactive semantics from humans canbe exploited and passed to objects to intensify interactions, b) interactivecorrelations among humans, objects and interactions are integrated to promotepredictions. Based on modules above, we construct an end-to-end trainableframework named Relation Reasoning Network (abbr. RR-Net). Extensiveexperiments show that our proposed RR-Net sets a new state-of-the-art on bothV-COCO and HICO-DET benchmarks and improves the baseline about 5.5% and 9.8%relatively, validating that this first effort in exploring relation reasoningand integrating interactive semantics has brought obvious improvement forend-to-end HOI detection.

RR-Net: Injecting Interactive Semantics in Human-Object Interaction Detection