Abstract

To investigate research questions overarching the domains of augmented, virtual and mixed reality as well as IoT- and web-based technologies with respect to Human-Computer Interaction tasks, the Ubi-Interact framework is developed at the IN-FAR research group.

This thesis deals with developing a Python software suite to use this framework, in a way that is tailored for use by researchers and students alike, no matter their domain specific backgrounds. It also showcases the advantages of the setup by providing a modular solution for common Optical Character Recognition tasks as a distributed computation module for Ubi-Interact, using the Tesseract engine combined with different image processing algorithms to achieve a robust and flexible solution with sufficient performance to be used in real time applications.

It is shown that Ubi-Interact excels at breaking down problems (or solutions) to different complexities, and allows flexible workflows – from rapid prototyping to performance oriented low level development – which is specifically highlighted by the features of the Python suite.

Results/Implementation/Project Description

Thesis repository

Documentation of Ubi Interact Python Node Package

Conclusion

For results and discussion of the OCR processing refer to chapter 3 of the attached thesis.