What is it about?
There are several gesture recognizers for 2D or 3D gestures, based on different techniques. We developed seven 3D gesture recognizers based on four 2D recognizers selected from the state-of-the-art to recognize trajectories, which represent sequential data of a single point. We implemented a testing framework and tested the recognizers on three datasets.
Featured Image
Photo by Ian Noble on Unsplash
Why is it important?
Our findings show that we can extend 2D stroke gesture recognizers to three dimensions, and we can interpret a 3D trajectory as three 2D trajectories projected on each plane. From the evaluation, the 3D extension of the 2D recognizer $P+ provides excellent results.
Perspectives
This article gives an idea on which recognition module to choose to recognize 3D trajectories, but also which conditions to have the most favourable results.
Mehdi Ousmer
Universite catholique de Louvain
Read the Original
This page is a summary of: Recognizing 3D Trajectories as 2D Multi-stroke Gestures, Proceedings of the ACM on Human-Computer Interaction, November 2020, ACM (Association for Computing Machinery),
DOI: 10.1145/3427326.
You can read the full text:
Contributors
The following have contributed to this page
