MR-GDINO: Efficient Open-World Continual Object Detection
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_ | 1866912167154941952 |
|---|---|
| author | Dong, Bowen Huang, Zitong Yang, Guanglei Zhang, Lei Zuo, Wangmeng |
| author_facet | Dong, Bowen Huang, Zitong Yang, Guanglei Zhang, Lei Zuo, Wangmeng |
| contents | Open-world (OW) recognition and detection models show strong zero- and few-shot adaptation abilities, inspiring their use as initializations in continual learning methods to improve performance. Despite promising results on seen classes, such OW abilities on unseen classes are largely degenerated due to catastrophic forgetting. To tackle this challenge, we propose an open-world continual object detection task, requiring detectors to generalize to old, new, and unseen categories in continual learning scenarios. Based on this task, we present a challenging yet practical OW-COD benchmark to assess detection abilities. The goal is to motivate OW detectors to simultaneously preserve learned classes, adapt to new classes, and maintain open-world capabilities under few-shot adaptations. To mitigate forgetting in unseen categories, we propose MR-GDINO, a strong, efficient and scalable baseline via memory and retrieval mechanisms within a highly scalable memory pool. Experimental results show that existing continual detectors suffer from severe forgetting for both seen and unseen categories. In contrast, MR-GDINO largely mitigates forgetting with only 0.1% activated extra parameters, achieving state-of-the-art performance for old, new, and unseen categories. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2412_15979 |
| institution | arXiv |
| publishDate | 2024 |
| record_format | arxiv |
| spellingShingle | MR-GDINO: Efficient Open-World Continual Object Detection Dong, Bowen Huang, Zitong Yang, Guanglei Zhang, Lei Zuo, Wangmeng Computer Vision and Pattern Recognition Open-world (OW) recognition and detection models show strong zero- and few-shot adaptation abilities, inspiring their use as initializations in continual learning methods to improve performance. Despite promising results on seen classes, such OW abilities on unseen classes are largely degenerated due to catastrophic forgetting. To tackle this challenge, we propose an open-world continual object detection task, requiring detectors to generalize to old, new, and unseen categories in continual learning scenarios. Based on this task, we present a challenging yet practical OW-COD benchmark to assess detection abilities. The goal is to motivate OW detectors to simultaneously preserve learned classes, adapt to new classes, and maintain open-world capabilities under few-shot adaptations. To mitigate forgetting in unseen categories, we propose MR-GDINO, a strong, efficient and scalable baseline via memory and retrieval mechanisms within a highly scalable memory pool. Experimental results show that existing continual detectors suffer from severe forgetting for both seen and unseen categories. In contrast, MR-GDINO largely mitigates forgetting with only 0.1% activated extra parameters, achieving state-of-the-art performance for old, new, and unseen categories. |
| title | MR-GDINO: Efficient Open-World Continual Object Detection |
| topic | Computer Vision and Pattern Recognition |
| url | https://arxiv.org/abs/2412.15979 |