JoVE - Journal of Visualized Experiments
JoVE Visualize
Contact Us

A dataset of exocentric images capturing hand gestures in multiplayer games

  • 0Department of Computer Science, American International University - Bangladesh, Kuratoli 408/1, Dhaka, Bangladesh.
Data in Brief +

|

November 10, 2025

|

November 10, 2025

View abstract on PubMed

Summary

This summary is machine-generated.

This dataset offers 7221 hand gesture images from multiplayer games like Ludo and Poker, captured under varied lighting. It aids research in computer vision and human-computer interaction for gesture recognition.

Area Of Science

  • Computer Vision
  • Human-Computer Interaction
  • Extended Reality (XR)

Background

  • Multiplayer tabletop games involve complex hand gestures and interactions.
  • Existing datasets may lack diverse conditions like varied lighting and multiple participants.
  • Realistic gesture data is crucial for advancing human-computer interaction.

Purpose Of The Study

  • To introduce a novel dataset of exocentric hand gestures from multiplayer tabletop games.
  • To provide a resource for developing and evaluating gesture recognition algorithms.
  • To support research in areas like hand-object interaction and occlusion handling.

Main Methods

  • Recorded 7221 exocentric frame images of hand gestures during Ludo, Poker, and Snakes and Ladders gameplay.
  • Used a Google Pixel 6a smartphone at 30 frames per second (FPS), extracting frames every four seconds.
  • Images were resized to 512x512 pixels and captured under four lighting conditions with 2-4 players.

Main Results

  • The dataset features authentic gesture dynamics, including motion blur, occlusions, and object manipulations.
  • It includes footage from varied lighting conditions (natural, white, yellow, dim) and multi-person sessions.
  • Consistent top-down viewpoints were maintained for all recordings.

Conclusions

  • This dataset facilitates advancements in gesture recognition for collaborative environments and interactive gaming.
  • It serves as a valuable resource for supervised and unsupervised learning tasks in computer vision.
  • Enables benchmarking of gesture-tracking algorithms and XR applications in realistic gaming scenarios.

Keywords: