Exploiting Gaussian Agnostic Representation Learning with Diffusion Priors for Enhanced Infrared Small Target Detection

2025-07-24

AI Summary

To address the robustness deficiency of infrared small target detection (ISTD) in real-world scenarios caused by scarce high-quality annotated data, this paper proposes a novel framework integrating Gaussian-agnostic representation learning with diffusion-based priors. Our method comprises two key components: (1) a Gaussian group compressor that performs non-uniform quantization to preserve critical structural details, and (2) a two-stage diffusion model—first introducing diffusion priors into synthetic sample reconstruction to better approximate the underlying distribution of real infrared data. Under data-constrained conditions, the framework significantly enhances model generalization and environmental adaptability. Experiments demonstrate superior synthetic sample fidelity and detection accuracy over state-of-the-art methods, achieving SOTA performance across multiple data-scarce settings. These results empirically validate the proposed method’s robustness and practical efficacy.

Technology Category

• Computer Vision: Diffusion Models for Vision
• Intelligent Robots: State Estimation
• Machine Learning: Calibration & Uncertainty Quantification

Application Category

• Search and Retrieval-Augmented AI: Web query analysis, representation and understanding
• Systems and Infrastructure for Web, Mobile and WoT: Applied ML and AI for Web-based mobile applications
• Web Mining and Content Analysis: Large pretrained models with web data

Abstract

Infrared small target detection (ISTD) plays a vital role in numerous practical applications. In pursuit of determining the performance boundaries, researchers employ large and expensive manual-labeling data for representation learning. Nevertheless, this approach renders the state-of-the-art ISTD methods highly fragile in real-world challenges. In this paper, we first study the variation in detection performance across several mainstream methods under various scarcity -- namely, the absence of high-quality infrared data -- that challenge the prevailing theories about practical ISTD. To address this concern, we introduce the Gaussian Agnostic Representation Learning. Specifically, we propose the Gaussian Group Squeezer, leveraging Gaussian sampling and compression for non-uniform quantization. By exploiting a diverse array of training samples, we enhance the resilience of ISTD models against various challenges. Then, we introduce two-stage diffusion models for real-world reconstruction. By aligning quantized signals closely with real-world distributions, we significantly elevate the quality and fidelity of the synthetic samples. Comparative evaluations against state-of-the-art detection methods in various scarcity scenarios demonstrate the efficacy of the proposed approach.

Problem

• Enhancing infrared small target detection with limited data
• Improving model resilience using Gaussian agnostic representation learning
• Elevating synthetic sample quality via diffusion-based reconstruction

Innovation

• Gaussian Agnostic Representation Learning for ISTD
• Gaussian Group Squeezer for non-uniform quantization
• Two-stage diffusion models for real-world reconstruction