DCPNet: Distribution Calibration Prototypical Network for Few-Shot Image Classification

Deep learning has witnessed significant advancements in various tasks and has displayed exceptional performance. However, traditional deep learning techniques often necessitate the utilization of extensive labeled data for training, a requirement that is challenging to fulfill in many real-world sce...

Full description

Saved in:
Bibliographic Details
Main Authors: Ranhui Xu, Kaizhong Jiang, Lulu Qi, Shaojie Zhao, Mingming Zheng
Format: Article
Language:English
Published: IEEE 2024-01-01
Series:IEEE Access
Subjects:
Online Access:https://ieeexplore.ieee.org/document/10522678/
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1850126419698384896
author Ranhui Xu
Kaizhong Jiang
Lulu Qi
Shaojie Zhao
Mingming Zheng
author_facet Ranhui Xu
Kaizhong Jiang
Lulu Qi
Shaojie Zhao
Mingming Zheng
author_sort Ranhui Xu
collection DOAJ
description Deep learning has witnessed significant advancements in various tasks and has displayed exceptional performance. However, traditional deep learning techniques often necessitate the utilization of extensive labeled data for training, a requirement that is challenging to fulfill in many real-world scenarios. This limitation has given rise to the field of few-shot learning (FSL). In this paper, we introduce a Distribution Calibration Prototypical Network (DCPNet), aiming to address the limitations of prototypical networks in terms of their weak feature extraction capabilities and the inability of their classifier boundaries to align with the dataset. DCPNet incorporates a parallel hierarchical feature extraction module and a few-shot differentiation loss function to fine-tune the metric learning for better feature representation. This approach employs a parallel approach to extract features based on the semantic depth of image hierarchical extraction and incorporates contrastive learning to achieve feature vector fusion. Furthermore, DCPNet incorporates an improved distribution calibration method that leverages information from the base class dataset to align classifier boundaries with the dataset. To validate our approach, we conducted comparative experiments on datasets such as Mini-Imagenet, Omniglot, and CUB using classical baseline methods. In additional, we conducted ablation experiments on the Mini-Imagenet to assess the performance effectiveness of each component of the model. The results demonstrate that the proposed method presented in this paper outperforms other approaches and offer new insights into the field of few-shot image classification.
format Article
id doaj-art-d15dd45018f6484284604c847c5a7828
institution OA Journals
issn 2169-3536
language English
publishDate 2024-01-01
publisher IEEE
record_format Article
series IEEE Access
spelling doaj-art-d15dd45018f6484284604c847c5a78282025-08-20T02:33:55ZengIEEEIEEE Access2169-35362024-01-0112670366704510.1109/ACCESS.2024.339813410522678DCPNet: Distribution Calibration Prototypical Network for Few-Shot Image ClassificationRanhui Xu0https://orcid.org/0009-0007-6463-1855Kaizhong Jiang1Lulu Qi2Shaojie Zhao3https://orcid.org/0000-0003-3836-7759Mingming Zheng4https://orcid.org/0009-0002-9084-0603School of Mathematics, Physics and Statistics, Shanghai University of Engineering Science, Shanghai, ChinaSchool of Mathematics, Physics and Statistics, Shanghai University of Engineering Science, Shanghai, ChinaSchool of Mathematics, Physics and Statistics, Shanghai University of Engineering Science, Shanghai, ChinaSchool of Mathematics, Physics and Statistics, Shanghai University of Engineering Science, Shanghai, ChinaSchool of Mathematics, Physics and Statistics, Shanghai University of Engineering Science, Shanghai, ChinaDeep learning has witnessed significant advancements in various tasks and has displayed exceptional performance. However, traditional deep learning techniques often necessitate the utilization of extensive labeled data for training, a requirement that is challenging to fulfill in many real-world scenarios. This limitation has given rise to the field of few-shot learning (FSL). In this paper, we introduce a Distribution Calibration Prototypical Network (DCPNet), aiming to address the limitations of prototypical networks in terms of their weak feature extraction capabilities and the inability of their classifier boundaries to align with the dataset. DCPNet incorporates a parallel hierarchical feature extraction module and a few-shot differentiation loss function to fine-tune the metric learning for better feature representation. This approach employs a parallel approach to extract features based on the semantic depth of image hierarchical extraction and incorporates contrastive learning to achieve feature vector fusion. Furthermore, DCPNet incorporates an improved distribution calibration method that leverages information from the base class dataset to align classifier boundaries with the dataset. To validate our approach, we conducted comparative experiments on datasets such as Mini-Imagenet, Omniglot, and CUB using classical baseline methods. In additional, we conducted ablation experiments on the Mini-Imagenet to assess the performance effectiveness of each component of the model. The results demonstrate that the proposed method presented in this paper outperforms other approaches and offer new insights into the field of few-shot image classification.https://ieeexplore.ieee.org/document/10522678/Improved distribution calibrationfew-shot learningprototypical networkimage classificationcomputer vision
spellingShingle Ranhui Xu
Kaizhong Jiang
Lulu Qi
Shaojie Zhao
Mingming Zheng
DCPNet: Distribution Calibration Prototypical Network for Few-Shot Image Classification
IEEE Access
Improved distribution calibration
few-shot learning
prototypical network
image classification
computer vision
title DCPNet: Distribution Calibration Prototypical Network for Few-Shot Image Classification
title_full DCPNet: Distribution Calibration Prototypical Network for Few-Shot Image Classification
title_fullStr DCPNet: Distribution Calibration Prototypical Network for Few-Shot Image Classification
title_full_unstemmed DCPNet: Distribution Calibration Prototypical Network for Few-Shot Image Classification
title_short DCPNet: Distribution Calibration Prototypical Network for Few-Shot Image Classification
title_sort dcpnet distribution calibration prototypical network for few shot image classification
topic Improved distribution calibration
few-shot learning
prototypical network
image classification
computer vision
url https://ieeexplore.ieee.org/document/10522678/
work_keys_str_mv AT ranhuixu dcpnetdistributioncalibrationprototypicalnetworkforfewshotimageclassification
AT kaizhongjiang dcpnetdistributioncalibrationprototypicalnetworkforfewshotimageclassification
AT luluqi dcpnetdistributioncalibrationprototypicalnetworkforfewshotimageclassification
AT shaojiezhao dcpnetdistributioncalibrationprototypicalnetworkforfewshotimageclassification
AT mingmingzheng dcpnetdistributioncalibrationprototypicalnetworkforfewshotimageclassification