North Rhine-Westphalia AI, Machine Learning and Computer Vision Meetup

How can robots learn to see without relying on massive labeled datasets? In this session, we’ll explore how adaptive robotic planning and semi-supervised active learning help robots improve their semantic vision in unknown environments. These approaches enable robots to decide what data to collect, when human labels are needed, and what can be learned autonomously from their understanding of the environment, reducing labeling effort while enhancing vision capabilities.

AI in agriculture presents unique challenges, from data curation to model evaluation. In this session, we’ll explore techniques for improving dataset quality, detecting anomalies, and leveraging embeddings to refine AI workflows. Using real-world coffee datasets from Colombia, we’ll dive into segmentation, similarity search, and feature extraction methods to enhance AI-driven decision-making. Whether you’re working with farm robotics, remote sensing, or plant phenotyping, this talk will provide practical insights into building more robust and reliable AI solutions. Let’s brew better AI together!