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

Person search in media has seen increasing potential in Internetapplications, such as video clipping and character collection. This task iscommon but overlooked by previous person search works which focus onsurveillance scenes. The media scenarios have some different challenges fromsurveillance scenes. For example, a person may change his clothes frequently.To alleviate this issue, this paper proposes a Unified Detector and GraphNetwork (UDGNet) for person search in media. UDGNet is the first person searchframework to detect and re-identify the human body and head simultaneously.Specifically, it first builds two branches based on a unified network to detectthe human body and head, then the detected body and head are used forre-identification. This dual-task approach can significantly enhancediscriminative learning. To tackle the cloth-changing issue, UDGNet builds twographs to explore reliable links among cloth-changing samples and utilizes agraph network to learn better embeddings. This design effectively enhances therobustness of person search to cloth-changing challenges. Besides, wedemonstrate that UDGNet can be implemented with both anchor-based andanchor-free person search frameworks and further achieve performanceimprovement. This paper also contributes a large-scale dataset for PersonSearch in Media (PSM), which provides both body and head annotations. It is byfar the largest dataset for person search in media. Experiments show thatUDGNet improves the anchor-free model AlignPS by 12.1% in mAP. Meanwhile, itshows good generalization across surveillance and longterm scenarios. Thedataset and code will be available at: https://github.com/shuxjweb/PSM.git.

Head and Body: Unified Detector and Graph Network for Person Search in Media