Few-shot target-driven instance detection based on open-vocabulary object detection models
Fuente:
arXiv
Enregistré dans:
| Auteurs principaux: | , , , |
|---|---|
| Format: | Preprint |
| Publié: |
2024
|
| Sujets: | |
| Accès en ligne: | |
| Tags: |
Ajouter un tag
Pas de tags, Soyez le premier à ajouter un tag!
|
| _version_ | 1866916447911936000 |
|---|---|
| author | Crulis, Ben Serres, Barthelemy De Runz, Cyril Venturini, Gilles |
| author_facet | Crulis, Ben Serres, Barthelemy De Runz, Cyril Venturini, Gilles |
| contents | 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. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2410_16028 |
| institution | arXiv |
| publishDate | 2024 |
| record_format | arxiv |
| spellingShingle | Few-shot target-driven instance detection based on open-vocabulary object detection models Crulis, Ben Serres, Barthelemy De Runz, Cyril Venturini, Gilles Computer Vision and Pattern Recognition 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. |
| title | Few-shot target-driven instance detection based on open-vocabulary object detection models |
| topic | Computer Vision and Pattern Recognition |
| url | https://arxiv.org/abs/2410.16028 |