Cross-Domain Few-Shot Segmentation via Ordinary Differential Equations over Time Intervals

2025-09-01

AI Summary

Existing cross-domain few-shot segmentation (CD-FSS) methods rely on loosely coupled, independent modules, impeding effective cross-domain knowledge transfer and limiting inter-module synergy. To address this, we propose FSS-TIs—a novel framework that, for the first time, models domain-invariant and domain-specific feature spectra as a continuous-time dynamical system governed by ordinary differential equations (ODEs), enabling joint optimization and smooth temporal evolution of features across domains. Our approach unifies feature-space exploration and target-domain distribution modeling as an ODE parameter learning problem, integrating Fourier spectral analysis, stochastic affine transformations with controlled perturbations, and a rigorously constrained support-sample selection strategy. Evaluated on two CD-FSS task settings comprising five benchmark datasets, FSS-TIs achieves significant improvements over state-of-the-art methods. Ablation studies confirm its strong generalization capability and demonstrate clear performance gains from module-level cooperation.

Technology Category

• Machine Learning: Transfer, Domain Adaptation, Multi-Task Learning
• Computer Vision: Segmentation
• Constraint Satisfaction and Optimization: Distributed CSP/Optimization

Application Category

• Search and Retrieval-Augmented AI: Vertical and domain-specific search
• Graph Algorithms and Modeling for the Web: Efficient manipulation of static and dynamic Web-related graphs
• User Modeling, Personalization and Recommendation: Federated recommendation systems and personalization

Abstract

Cross-domain few-shot segmentation (CD-FSS) not only enables the segmentation of unseen categories with very limited samples, but also improves cross-domain generalization ability within the few-shot segmentation framework. Currently, existing CD-FSS studies typically design multiple independent modules to enhance the cross-domain generalization ability of feature representations. However, the independence among these modules hinders the effective flow of knowledge, making it difficult to fully leverage their collective potential. In contrast, this paper proposes an all-in-one module based on ordinary differential equations and Fourier transform, resulting in a structurally concise method--Few-Shot Segmentation over Time Intervals (FSS-TIs). FSS-TIs assumes the existence of an ODE relationship between the spectra (including amplitude and phase spectra) of domain-specific features and domain-agnostic features. This ODE formulation yields an iterative transformation process along a sequence of time intervals, while simultaneously applying affine transformations with randomized perturbations to the spectra. In doing so, the exploration of domain-agnostic feature representation spaces and the simulation of diverse potential target-domain distributions are reformulated as an optimization process over the intrinsic parameters of the ODE. Moreover, we strictly constrain the support-sample selection during target-domain fine-tuning so that it is consistent with the requirements of real-world few-shot segmentation tasks. For evaluation, we introduce five datasets from substantially different domains and define two sets of cross-domain few-shot segmentation tasks to comprehensively analyze the performance of FSS-TIs. Experimental results demonstrate the superiority of FSS-TIs over existing CD-FSS methods, and in-depth ablation studies further validate the cross-domain adaptability of FSS-TIs.

Problem

• Enhancing cross-domain generalization in few-shot image segmentation
• Overcoming knowledge flow barriers between independent modules
• Modeling domain-agnostic features through ODE-based spectral transformations

Innovation

• ODE-based iterative transformation for domain-agnostic features
• Fourier transform with randomized spectral perturbations
• Constrained support-sample selection for real-world tasks