Annotation tools for computer vision tasks
Christos Moschidis, Εleni Vrochidou, George A. Papakostas · 2025
Common computer vision (CV) tasks include image classification, object detection, segmentation, and recognition. To handle such tasks, machine learning (ML) models for image processing require a great amount of annotated training data. While datasets are expanding in size and variety, annotation becomes demanding, since its quality can severely affect the models’ performance. Thus, several annotation tools have been developed and designated for specific applications and model requirements. This work aims to provide an overview of the most up-to-date annotation tools for computer vision tasks, including 2D and 3D image data and video, comparatively highlighting their advantages and limitations. The appropriateness of each tool for specific tasks is emphasized, providing a reference map for researchers towards determining the annotation tool best tailored to their needs. Future trends in image annotation are also discussed.