Current large open vision models could be useful for one and few-shot object recognition. Nevertheless, gradient-based re-training solutions are costly. On the other hand, open-vocabulary object detection models bring closer visual and textual concepts in the same latent space, allowing zero-shot detection via prompting at small computational cost. We propose a lightweight method to turn the latter into a one-shot or few-shot object recognition models without requiring textual descriptions. Our experiments on the TEgO dataset using the YOLO-World model as a base show that performance increases with the model size, the number of examples and the use of image augmentation.
@article{arxiv.2410.16028,
title = {Few-shot target-driven instance detection based on open-vocabulary object detection models},
author = {Ben Crulis and Barthelemy Serres and Cyril De Runz and Gilles Venturini},
journal= {arXiv preprint arXiv:2410.16028},
year = {2024}
}