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 Tools annual

Global: Zendy Free Plan

Global: Zendy Tools monthly

Global: Zendy Plus monthly

Video-based human pose estimation (VHPE) is a vital yet challenging task.While deep learning methods have made significant progress for the VHPE, mostapproaches to this task implicitly model the long-range interaction betweenjoints by enlarging the receptive field of the convolution. Unlike priormethods, we design a lightweight and plug-and-play joint relation extractor(JRE) to model the associative relationship between joints explicitly andautomatically. The JRE takes the pseudo heatmaps of joints as input andcalculates the similarity between pseudo heatmaps. In this way, the JREflexibly learns the relationship between any two joints, allowing it to learnthe rich spatial configuration of human poses. Moreover, the JRE can inferinvisible joints according to the relationship between joints, which isbeneficial for the model to locate occluded joints. Then, combined withtemporal semantic continuity modeling, we propose a Relation-based PoseSemantics Transfer Network (RPSTN) for video-based human pose estimation.Specifically, to capture the temporal dynamics of poses, the pose semanticinformation of the current frame is transferred to the next with a jointrelation guided pose semantics propagator (JRPSP). The proposed model cantransfer the pose semantic features from the non-occluded frame to the occludedframe, making our method robust to the occlusion. Furthermore, the proposed JREmodule is also suitable for image-based human pose estimation. The proposedRPSTN achieves state-of-the-art results on the video-based Penn Action dataset,Sub-JHMDB dataset, and PoseTrack2018 dataset. Moreover, the proposed JREimproves the performance of backbones on the image-based COCO2017 dataset. Codeis available at https://github.com/YHDang/pose-estimation.

Relation-Based Associative Joint Location for Human Pose Estimation in Videos