4D Feet: Registering Walking Foot Shapes Using Attention Enhanced Dynamic-Synchronized Graph Convolutional LSTM Network

Farzam Tajdari, Toon Huysmans, Xinhe Yao, Jun Xu, Maryam Zebarjadi, Yu Song

Research output: Contribution to journalArticleScientificpeer-review

1 Citation (Scopus)
16 Downloads (Pure)

Abstract

4D-scans of dynamic deformable human body parts help researchers have a better understanding of spatiotemporal features. However, reconstructing 4D-scans utilizing multiple asynchronous cameras encounters two main challenges: 1) finding dynamic correspondences among different frames captured by each camera at the timestamps of the camera in terms of dynamic feature recognition, and 2) reconstructing 3D-shapes from the combined point clouds captured by different cameras at asynchronous timestamps in terms of multi-view fusion. Here, we introduce a generic framework able to 1) find and align dynamic features in the 3D-scans captured by each camera using the nonrigid-iterative-closest-farthestpoints algorithm; 2) synchronize scans captured by asynchronous cameras through a novel ADGC-LSTMbased-network capable of aligning 3D-scans captured by different cameras to the timeline of a specific camera; and 3) register a high-quality template to synchronized scans at each timestamp to form a highquality 3D-mesh model using a non-rigid registration method. With a newly developed 4D-foot-scanner, we validate the framework and create the first open-access data-set, namely the 4D-feet. It includes 4Dshapes (15 fps) of the right and left feet of 58 participants (116 feet including 5147 3D-frames), covering significant phases of the gait cycle. The results demonstrate the effectiveness of the proposed framework, especially in synchronizing asynchronous 4D-scans.

Original languageEnglish
Pages (from-to)343-355
Number of pages13
JournalIEEE Open Journal of the Computer Society
Volume5
DOIs
Publication statusPublished - 2024

Keywords

  • 4D foot scanner
  • Cameras
  • dynamic feature recognition
  • Feature extraction
  • Foot
  • LSTM network
  • nonrigid registration
  • Point cloud compression
  • Shape
  • Synchronization
  • synchronized scans
  • Three-dimensional displays

Fingerprint

Dive into the research topics of '4D Feet: Registering Walking Foot Shapes Using Attention Enhanced Dynamic-Synchronized Graph Convolutional LSTM Network'. Together they form a unique fingerprint.

Cite this