Ivan-ISTD: Rethinking Cross-domain Heteroscedastic Noise Perturbations in Infrared Small Target Detection

2025-10-14

AI Summary

To address the concurrent challenges of cross-domain shift and heteroscedastic noise in infrared small target detection (ISTD), this paper proposes a dual-waveguide invariance learning framework. First, it introduces a wavelet-guided cross-domain sample synthesis mechanism to mitigate inter-domain distribution discrepancies. Second, it establishes a self-supervised invariance learning paradigm grounded in dynamic modeling of real-domain noise, overcoming distributional biases inherent in synthetic noise assumptions. Key contributions include: (1) Dynamic-ISTD—the first ISTD benchmark explicitly designed for cross-domain dynamic degradation; (2) a multi-band wavelet filtering and waveguide-based cross-domain synthesis technique; and (3) an end-to-end architecture integrating real-noise feature extraction with invariance learning. Extensive experiments on multiple real-world datasets demonstrate significant improvements over state-of-the-art methods, with enhanced cross-domain generalization and robustness against heterogeneous noise.

Technology Category

• Machine Learning: Transfer, Domain Adaptation, Multi-Task Learning
• Intelligent Robots: Multimodal Perception & Sensor Fusion
• Computer Vision: Adversarial Attacks & Robustness

Application Category

• Systems and Infrastructure for Web, Mobile and WoT: Applied ML and AI for Web-based mobile applications
• 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

Abstract

In the multimedia domain, Infrared Small Target Detection (ISTD) plays a important role in drone-based multi-modality sensing. To address the dual challenges of cross-domain shift and heteroscedastic noise perturbations in ISTD, we propose a doubly wavelet-guided Invariance learning framework(Ivan-ISTD). In the first stage, we generate training samples aligned with the target domain using Wavelet-guided Cross-domain Synthesis. This wavelet-guided alignment machine accurately separates the target background through multi-frequency wavelet filtering. In the second stage, we introduce Real-domain Noise Invariance Learning, which extracts real noise characteristics from the target domain to build a dynamic noise library. The model learns noise invariance through self-supervised loss, thereby overcoming the limitations of distribution bias in traditional artificial noise modeling. Finally, we create the Dynamic-ISTD Benchmark, a cross-domain dynamic degradation dataset that simulates the distribution shifts encountered in real-world applications. Additionally, we validate the versatility of our method using other real-world datasets. Experimental results demonstrate that our approach outperforms existing state-of-the-art methods in terms of many quantitative metrics. In particular, Ivan-ISTD demonstrates excellent robustness in cross-domain scenarios. The code for this work can be found at: https://github.com/nanjin1/Ivan-ISTD.

Problem

• Addressing cross-domain shift and heteroscedastic noise in infrared target detection
• Overcoming distribution bias in traditional artificial noise modeling
• Solving real-world infrared target detection with dynamic degradation scenarios

Innovation

• Wavelet-guided cross-domain synthesis for training samples
• Real-domain noise invariance learning with dynamic library
• Dynamic-ISTD benchmark simulating real-world distribution shifts