Hyperbolic Insights With Knowledge Distillation for Cross-Domain Few-Shot Learning
IEEE Transactions on Image Processing · 2025 · 34 · 1921-1933
DOI: 10.1109/tip.2025.3551647 ↗Kurz gefasst
Fachartikel in der Zeitschrift IEEE Transactions on Image Processing, 2025. Verfasst von Xi Yang und 3 weiteren Autor:innen. Die Arbeit wurde bisher 11-mal zitiert (Crossref) und nennt 67 Literaturangaben. Gefördert von: National Natural Science Foundation of China, Key Research and Development Program of Shaanxi, Shaanxi Outstanding Youth Science Fund Project.
Abstract (Originalsprache)
Cross-domain few-shot learning aims to achieve swift generalization between a source domain and a target domain using a limited number of images. Current research predominantly relies on generalized feature embeddings, employing metric classifiers in Euclidean space for classification. However, due to existing disparities among different data domains, attaining generalized features in the embedding becomes challenging. Additionally, the rise in data domains leads to high-dimensional Euclidean spaces. To address the above problems, we introduce a cross-domain few-shot learning method named Hyperbolic Insights with Knowledge Distillation (HIKD). By integrating knowledge distillation, it enhances the model's generalization performance, thereby significantly improving task performance. Hyperbolic space, in comparison to Euclidean space, offers a larger capacity and supports the learning of hierarchical structures among images, which can aid generalized learning across different data domains. So we map the Euclidean space features to the hyperbolic space via hyperbolic embedding and utilize hyperbolic fitting distillation method in the meta-training phase to obtain multi-domain unified generalization representation. In the meta-testing phase, accounting for biases between the source and target domains, we present a hyperbolic adaptive module to adjust embedded features and eliminate inter-domain gap. Experiments on the Meta-Dataset demonstrate that HIKD outperforms state-of-the-arts methods with the average accuracy of 80.6%.
Förderung
- National Natural Science Foundation of China
- Key Research and Development Program of Shaanxi
- Shaanxi Outstanding Youth Science Fund Project
- Shaanxi Province Core Technology Research and Development Project
- CCF-Baidu Open Fund; and in part by the Fundamental Research Funds for the Central Universities
Zitieren
APA
Yang, X., Kong, D., Wang, N., & Gao, X. (2025). Hyperbolic Insights With Knowledge Distillation for Cross-Domain Few-Shot Learning. IEEE Transactions on Image Processing, 34, 1921-1933. https://doi.org/10.1109/tip.2025.3551647
Vancouver
Yang X, Kong D, Wang N, Gao X. Hyperbolic Insights With Knowledge Distillation for Cross-Domain Few-Shot Learning. IEEE Transactions on Image Processing. 2025;34:1921-1933. doi:10.1109/tip.2025.3551647
BibTeX
@article{Yang2025,
author = {Yang, Xi and Kong, Dechen and Wang, Nannan and Gao, Xinbo},
title = {Hyperbolic Insights With Knowledge Distillation for Cross-Domain Few-Shot Learning},
journal = {IEEE Transactions on Image Processing},
year = {2025},
volume = {34},
pages = {1921--1933},
doi = {10.1109/tip.2025.3551647}
}
EUVIMED