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 body orientation estimation (HBOE) is widely applied into variousapplications, including robotics, surveillance, pedestrian analysis andautonomous driving. Although many approaches have been addressing the HBOEproblem from specific under-controlled scenes to challenging in-the-wildenvironments, they assume human instances are already detected and take a wellcropped sub-image as the input. This setting is less efficient and prone toerrors in real application, such as crowds of people. In the paper, we proposea single-stage end-to-end trainable framework for tackling the HBOE problemwith multi-persons. By integrating the prediction of bounding boxes anddirection angles in one embedding, our method can jointly estimate the locationand orientation of all bodies in one image directly. Our key idea is tointegrate the HBOE task into the multi-scale anchor channel predictions ofpersons for concurrently benefiting from engaged intermediate features.Therefore, our approach can naturally adapt to difficult instances involvinglow resolution and occlusion as in object detection. We validated theefficiency and effectiveness of our method in the recently presented benchmarkMEBOW with extensive experiments. Besides, we completed ambiguous instancesignored by the MEBOW dataset, and provided corresponding weak body-orientationlabels to keep the integrity and consistency of it for supporting studiestoward multi-persons. Our work is available athttps://github.com/hnuzhy/JointBDOE.

Joint Multi-Person Body Detection and Orientation Estimation via One Unified Embedding