Meta-Adapter: An Online Few-shot Learner for Vision-Language Model

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Cheng, Cheng, Song, Lin, Xue, Ruoyi, Wang, Hang, Sun, Hongbin, Ge, Yixiao, Shan, Ying
Format: Preprint
Published: 2023
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866917563685928960
author Cheng, Cheng
Song, Lin
Xue, Ruoyi
Wang, Hang
Sun, Hongbin
Ge, Yixiao
Shan, Ying
author_facet Cheng, Cheng
Song, Lin
Xue, Ruoyi
Wang, Hang
Sun, Hongbin
Ge, Yixiao
Shan, Ying
contents The contrastive vision-language pre-training, known as CLIP, demonstrates remarkable potential in perceiving open-world visual concepts, enabling effective zero-shot image recognition. Nevertheless, few-shot learning methods based on CLIP typically require offline fine-tuning of the parameters on few-shot samples, resulting in longer inference time and the risk of over-fitting in certain domains. To tackle these challenges, we propose the Meta-Adapter, a lightweight residual-style adapter, to refine the CLIP features guided by the few-shot samples in an online manner. With a few training samples, our method can enable effective few-shot learning capabilities and generalize to unseen data or tasks without additional fine-tuning, achieving competitive performance and high efficiency. Without bells and whistles, our approach outperforms the state-of-the-art online few-shot learning method by an average of 3.6\% on eight image classification datasets with higher inference speed. Furthermore, our model is simple and flexible, serving as a plug-and-play module directly applicable to downstream tasks. Without further fine-tuning, Meta-Adapter obtains notable performance improvements in open-vocabulary object detection and segmentation tasks.
format Preprint
id arxiv_https___arxiv_org_abs_2311_03774
institution arXiv
publishDate 2023
record_format arxiv
spellingShingle Meta-Adapter: An Online Few-shot Learner for Vision-Language Model
Cheng, Cheng
Song, Lin
Xue, Ruoyi
Wang, Hang
Sun, Hongbin
Ge, Yixiao
Shan, Ying
Computer Vision and Pattern Recognition
The contrastive vision-language pre-training, known as CLIP, demonstrates remarkable potential in perceiving open-world visual concepts, enabling effective zero-shot image recognition. Nevertheless, few-shot learning methods based on CLIP typically require offline fine-tuning of the parameters on few-shot samples, resulting in longer inference time and the risk of over-fitting in certain domains. To tackle these challenges, we propose the Meta-Adapter, a lightweight residual-style adapter, to refine the CLIP features guided by the few-shot samples in an online manner. With a few training samples, our method can enable effective few-shot learning capabilities and generalize to unseen data or tasks without additional fine-tuning, achieving competitive performance and high efficiency. Without bells and whistles, our approach outperforms the state-of-the-art online few-shot learning method by an average of 3.6\% on eight image classification datasets with higher inference speed. Furthermore, our model is simple and flexible, serving as a plug-and-play module directly applicable to downstream tasks. Without further fine-tuning, Meta-Adapter obtains notable performance improvements in open-vocabulary object detection and segmentation tasks.
title Meta-Adapter: An Online Few-shot Learner for Vision-Language Model
topic Computer Vision and Pattern Recognition
url https://arxiv.org/abs/2311.03774