Official implementation of: "Leveraging Vision-Language Models for Open-Vocabulary Instance Segmentation and Tracking" by Pätzold, Nogga & Behnke. Robotics and Automation Letters. 2025. - View it on GitHub