[NAVER2020] Hand gesture recognition


  • PI: Dr. Dang Tuan Linh
  • Time: 1/2021-12/2021
  • Funding: Naver company, South Korea
  • Main areas: Computer Vision, Deep Learning


The purpose of this project is to create a new dataset for different gestures using both hands. The self-built dataset, which consists of both static hand gestures and dynamic hand gestures, has different subjects (males, females), different backgrounds, different camera angles, more gesture numbers than the conventional datasets…

In addition, different state-of-the-art algorithms will also be investigated and customized to propose a suitable approach for our dataset.  This research not only focuses on normal models but also lightweight models which can be implemented on power and resources-constrained devices.