Kaya ter Burg is a researcher on multi-modal unterwater computer vision for litter detection. Her research focuses on advancing the project's image recognition system. She holds an academic background in Artificial Intelligence, with specialization in Computer Vision, Deep Learning, and Explainable AI.

In her talk, she will present the AI-based methods they have developed for the detection and classification of different types of litter as well as underwater flora and fauna from underwater camera and sonar images. The different challenges arising from the shallow underwater environment are discussed (such as turbidity, reflections, distortions, lack of labeled data) and some methods are presented to address them (including state-of-the-art algorithms such as YOLOv12). In SeaClear2.0, they further improve the camera-based object detection pipeline developed during SeaClear and extend this with additional information from a sonar. She will discuss the process of developing the litter detection pipeline for SeaClear2.0, including data collection, data processing, sensor fusion, and (multi-modal) object detection.