Analysis 3D Trajectories towards Evaluation of Mid-Air-Writing Recognition Systems

Writing a linguistic character or word in free space with a finger, marker, or handheld device is referred to as trajectory-based writing. It is widely used in situations where traditional pen-up and pen-down writing systems are inconvenient. It has a significant advantage over the gesture-based system due to its simple writing style. However, it is a difficult task due to the Characters that aren’t all the same and writing styles that aren’t all the same and the various systems used the nearest neighbor and root point interpretation were used to improve feature selection of trajectory. In this comparative study, mid air-writing recognition system using three-dimensional (3D) trajectories obtained by an image sensor that detects the fingertip works covered in detail. Moreover, the paper highlighted the different traditional works related to 3D trajectory based mid-air writing recognition systems and the algorithms used in those systems. An extensive comparative study was presented in this paper, revealed a dearth of information regarding mid air handwritten character recognition.