Luistervinq 2 – Activity recognition in nature using sound and AI
Problem Statement:
Luistervinq aims to provide insight in the use of outdoor environments. For example, do most people use a specific trail for hiking, jogging, or mountain biking? Governments, park owners, and nature conservers need to know how the natural environment is used by people so that the management for that area can be optimized. Due to privacy regulations, it is not possible to use cameras in public areas. Therefore, sound is used to automatically recognize the activities.
Task:
Investigate state of the art AI algorithms to classify activity from sound. Deploy algorithm on edge device. Evaluate performance versus power usage. Identify open issues.
Work:
20% Theory, 60% Implementation, 20% Writing
Contact:
Jacob Kamminga (j.w.kamminga@utwente.nl)