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

Human-centric perception (e.g. pedetrian detection, segmentation, poseestimation, and attribute analysis) is a long-standing problem for computervision. This paper introduces a unified and versatile framework (HQNet) forsingle-stage multi-person multi-task human-centric perception (HCP). Ourapproach centers on learning a unified human query representation, denoted asHuman Query, which captures intricate instance-level features for individualpersons and disentangles complex multi-person scenarios. Although different HCPtasks have been well-studied individually, single-stage multi-task learning ofHCP tasks has not been fully exploited in the literature due to the absence ofa comprehensive benchmark dataset. To address this gap, we proposeCOCO-UniHuman benchmark dataset to enable model development and comprehensiveevaluation. Experimental results demonstrate the proposed method'sstate-of-the-art performance among multi-task HCP models and its competitiveperformance compared to task-specific HCP models. Moreover, our experimentsunderscore Human Query's adaptability to new HCP tasks, thus demonstrating itsrobust generalization capability. Codes and data will be publicly accessible.

You Only Learn One Query: Learning Unified Human Query for Single-Stage Multi-Person Multi-Task Human-Centric Perception