ViT-VS is a visual servoing approach that leverages pretrained vision transformers for semantic feature extraction. Our framework combines the advantages of classical and learning-based visual ...