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

In human-computer interaction, it is important to accurately estimate thehand pose especially fingertips. However, traditional approaches for fingertiplocalization mainly rely on depth images and thus suffer considerably from thenoise and missing values. Instead of depth images, stereo images can alsoprovide 3D information of hands and promote 3D hand pose estimation. There arenevertheless limitations on the dataset size, global viewpoints, handarticulations and hand shapes in the publicly available stereo-based hand posedatasets. To mitigate these limitations and promote further research on handpose estimation from stereo images, we propose a new large-scale binocular handpose dataset called THU-Bi-Hand, offering a new perspective for fingertiplocalization. In the THU-Bi-Hand dataset, there are 447k pairs of stereo imagesof different hand shapes from 10 subjects with accurate 3D location annotationsof the wrist and five fingertips. Captured with minimal restriction on therange of hand motion, the dataset covers large global viewpoint space and handarticulation space. To better present the performance of fingertip localizationon THU-Bi-Hand, we propose a novel scheme termed Bi-stream Pose Guided RegionEnsemble Network (Bi-Pose-REN). It extracts more representative feature regionsaround joint points in the feature maps under the guidance of the previouslyestimated pose. The feature regions are integrated hierarchically according tothe topology of hand joints to regress the refined hand pose. Bi-Pose-REN andseveral existing methods are evaluated on THU-Bi-Hand so that benchmarks areprovided for further research. Experimental results show that our new methodhas achieved the best performance on THU-Bi-Hand.

Bi-stream Pose Guided Region Ensemble Network for Fingertip Localization from Stereo Images