SeqCSIST: Sequential Closely-Spaced Infrared Small Target Unmixing

2025-07-13

AI Summary

Distant, dense infrared small targets (CSISTs) often merge into subpixel-scale mixed blobs within a single frame due to optical diffraction and detector resolution limits, rendering them indistinguishable. Method: This paper proposes the first multi-frame subpixel unmixing framework, DeRefNet—a model-driven network incorporating a deformable temporal feature alignment (TDFA) module to enable adaptive inter-frame feature aggregation and precise subpixel localization. Contribution/Results: We introduce SeqCSIST—the first open-source benchmark dataset for sequential CSIST detection—along with a standardized evaluation toolkit to foster community-wide benchmarking. On our proprietary dataset, DeRefNet achieves a 5.3% mAP improvement over state-of-the-art methods, significantly enhancing separation accuracy for densely packed small targets.

Technology Category

• Computer Vision: Segmentation
• Constraint Satisfaction and Optimization: Distributed CSP/Optimization
• Machine Learning: Multi-instance/Multi-view Learning

Application Category

• Graph Algorithms and Modeling for the Web: Representation, reconstruction, and subgraph or motif discovery in Web-related graphs
• Search and Retrieval-Augmented AI: Web learning to rank, online learning, and counterfactual learning for ranking
• Systems and Infrastructure for Web, Mobile and WoT: Applied ML and AI for Web-based mobile applications

Abstract

Due to the limitation of the optical lens focal length and the resolution of the infrared detector, distant Closely-Spaced Infrared Small Target (CSIST) groups typically appear as mixing spots in the infrared image. In this paper, we propose a novel task, Sequential CSIST Unmixing, namely detecting all targets in the form of sub-pixel localization from a highly dense CSIST group. However, achieving such precise detection is an extremely difficult challenge. In addition, the lack of high-quality public datasets has also restricted the research progress. To this end, firstly, we contribute an open-source ecosystem, including SeqCSIST, a sequential benchmark dataset, and a toolkit that provides objective evaluation metrics for this special task, along with the implementation of 23 relevant methods. Furthermore, we propose the Deformable Refinement Network (DeRefNet), a model-driven deep learning framework that introduces a Temporal Deformable Feature Alignment (TDFA) module enabling adaptive inter-frame information aggregation. To the best of our knowledge, this work is the first endeavor to address the CSIST Unmixing task within a multi-frame paradigm. Experiments on the SeqCSIST dataset demonstrate that our method outperforms the state-of-the-art approaches with mean Average Precision (mAP) metric improved by 5.3%. Our dataset and toolkit are available from https://github.com/GrokCV/SeqCSIST.

Problem

• Detect sub-pixel targets in dense infrared groups
• Address lack of public datasets for CSIST research
• Improve precision in sequential CSIST unmixing task

Innovation

• Deformable Refinement Network for target detection
• Temporal Deformable Feature Alignment module
• Open-source SeqCSIST dataset and toolkit