Deep learning approaches to object detection have achieved reliable detection of specific object classes in images. However, extending a model’s detection capability to new object classes requires large amounts of annotated training data, which is costly and time-consuming to acquire, especially for long-tailed classes with insufficient representation in existing datasets. We compare four different methods of generating synthetic data to finetune object detection models on novel object categories, particularly when limited data is available in the form of object-centric data (multi-view images or 3D models). Our approaches are based on simple image processing techniques, 3D rendering, and image generation models, each varying in complexity and image realism. We assess the ability of models finetuned using synthetic data to generalize to novel object classes in real-world data, and we achieve significant performance boosts in our data-constrained experimental setting.