avaTTAR: Table Tennis Stroke Training with On-body and Detached Visualization in Augmented Reality

TL;DR

A new augmented reality (AR) system, avaTTAR, for table tennis stroke training that provides both “on-body” and “detached” visual cues, enabling users to visualize target strokes and correct their attempts effectively with this dual perspectives setup.

Abstract

Table tennis stroke training is a critical aspect of player development. We designed a new augmented reality (AR) system, avaTTAR, for table tennis stroke training. The system provides both "on-body" (first-person view) and "detached" (third-person view) visual cues, enabling users to visualize target strokes and correct their attempts effectively with this dual perspectives setup. By employing a combination of pose estimation algorithms and IMU sensors, avaTTAR captures and reconstructs the 3D body pose and paddle orientation of users during practice, allowing real-time comparison with expert strokes. Through a user study, we affirm avaTTAR's capacity to amplify player experience and training results.

Figures (11)

• Figure 1: Illustration of design requirements and system components. We have extracted four key design requirements from our formative interviews with individuals experienced in table tennis training. In response, we have elaborated on these requirements, resulting in the development of nine detailed modules for implementation.
• Figure 2: System's backend workflow. (a) The authoring backend: enables the expert to record and edit their movement posture data. (b) The learning backend: captures the user's motion and compares it with expert data, providing visual cues for stroke learning, and allows the user to adjust these visuals via the AR interface.
• Figure 3: The authoring interface: (a) The record mode allows the user to record video and stroke. (b) The edit mode interface allows authors to trim video and specify keyframes.
• Figure 4: (a) Our motion capture module consists of one webcam placed in front of the user and one IMU attached to the table tennis paddle. The webcam provides the visual input for the system and the IMU provides the paddle orientation. (b) Examples of motion capture results with different body and paddle postures.
• Figure 5: Showcase of our visualization. On-body cues: (a-1) movement trajectory; (a-2) paddle shadow shows paddle trajectory; (a-3) footwork trajectory. Detached cues: (b-1) expert avatar shows correct movement; (b-2) user avatar shows user's movement and highlights incorrect skeleton (in pink).