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

3D human pose estimation in outdoor environments has garnered increasingattention recently. However, prevalent 3D human pose datasets pertaining tooutdoor scenes lack diversity, as they predominantly utilize only one type ofmodality (RGB image or pointcloud), and often feature only one individualwithin each scene. This limited scope of dataset infrastructure considerablyhinders the variability of available data. In this article, we proposeHuman-M3, an outdoor multi-modal multi-view multi-person human pose databasewhich includes not only multi-view RGB videos of outdoor scenes but alsocorresponding pointclouds. In order to obtain accurate human poses, we proposean algorithm based on multi-modal data input to generate ground truthannotation. This benefits from robust pointcloud detection and tracking, whichsolves the problem of inaccurate human localization and matching ambiguity thatmay exist in previous multi-view RGB videos in outdoor multi-person scenes, andgenerates reliable ground truth annotations. Evaluation of multiple differentmodalities algorithms has shown that this database is challenging and suitablefor future research. Furthermore, we propose a 3D human pose estimationalgorithm based on multi-modal data input, which demonstrates the advantages ofmulti-modal data input for 3D human pose estimation. Code and data will bereleased on https://github.com/soullessrobot/Human-M3-Dataset.

Human-M3: A Multi-view Multi-modal Dataset for 3D Human Pose Estimation in Outdoor Scenes