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

Hand pose estimation (HPE) is a task that predicts and describes the handposes from images or video frames. When HPE models estimate hand poses capturedin a laboratory or under controlled environments, they normally deliver goodperformance. However, the real-world environment is complex, and variousuncertainties may happen, which could degrade the performance of HPE models.For example, the hands could be occluded, the visibility of hands could bereduced by imperfect exposure rate, and the contour of hands prone to beblurred during fast hand movements. In this work, we adopt metamorphic testingto evaluate the robustness of HPE models and provide suggestions on the choiceof HPE models for different applications. The robustness evaluation wasconducted on four state-of-the-art models, namely MediaPipe hands, OpenPose,BodyHands, and NSRM hand. We found that on average more than 80\% of the handscould not be identified by BodyHands, and at least 50\% of hands could not beidentified by MediaPipe hands when diagonal motion blur is introduced, while anaverage of more than 50\% of strongly underexposed hands could not be correctlyestimated by NSRM hand. Similarly, applying occlusions on only four hand jointswill also largely degrade the performance of these models. The experimentalresults show that occlusions, illumination variations, and motion blur are themain obstacles to the performance of existing HPE models. These findings maypave the way for researchers to improve the performance and robustness of handpose estimation models and their applications.

Robustness Evaluation in Hand Pose Estimation Models using Metamorphic Testing