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

Modern deep learning-based 3D pose estimation approaches require plenty of 3Dpose annotations. However, existing 3D datasets lack diversity, which limitsthe performance of current methods and their generalization ability. Althoughexisting methods utilize 2D pose annotations to help 3D pose estimation, theymainly focus on extracting 2D structural constraints from 2D poses, ignoringthe 3D information hidden in the images. In this paper, we propose a novelmethod to extract weak 3D information directly from 2D images without 3D posesupervision. Firstly, we utilize 2D pose annotations and perspective priorknowledge to generate the relationship of that keypoint is closer or fartherfrom the camera, called relative depth. We collect a 2D pose dataset (MCPC) andgenerate relative depth labels. Based on MCPC, we propose a weakly-supervisedpre-training (WSP) strategy to distinguish the depth relationship between twopoints in an image. WSP enables the learning of the relative depth of twokeypoints on lots of in-the-wild images, which is more capable of predictingdepth and generalization ability for 3D human pose estimation. Afterfine-tuning on 3D pose datasets, WSP achieves state-of-the-art results on twowidely-used benchmarks.

Weakly-supervised Pre-training for 3D Human Pose Estimation via Perspective Knowledge