发表机构
Nankai University; Dalian University of Technology; National University of Defense Technology（南开大学; 大连理工大学; 国防科技大学）
机构由 AI 辅助整理，请以论文原文为准。AI 中文总结
针对远距离红外密集小目标解混中盲分离与网格离散化的范式缺陷，提出信息性与连续性解混框架，构建CSIST-100K生态，并设计DISTA-Net++，以计数引导先验与连续坐标校正实现高效高精度解混。
AI 中文摘要
远距离红外成像经常面临密集目标簇，其衍射受限特征合并成一个不可区分的模糊斑点，掩盖了底层源的数量、亚像素位置和辐射强度。虽然深度学习推动了通用目标检测的进步，但由于系统性的基础设施缺失和根本性的范式不匹配，解决此类密集红外小目标（CSIST）的问题在很大程度上仍未得到探索。主导的公式将解混简化为盲目的、离散的亚像素分离，本质上是不充分的：没有语义引导，不适定的逆问题会产生模糊的解，饱受误检和漏检之苦，而基于网格的离散化将预测锁定在固定的晶格中心，使精度受制于极其昂贵的网格细化。我们认为，CSIST解混应具有信息性和连续性。为了夯实这一范式转变，我们为该领域建立了首个全面的开源生态系统，包括大规模CSIST-100K基准、定制的度量套件和GrokCSO工具包。在此基础之上，我们提出了DISTA-Net++，它锚定了一个动态深度展开主干，并配备了两个协同机制：一个计数引导先验，将全局目标计数作为显式语义约束注入以正则化解空间，以及一个连续坐标校正，通过回归离网格偏移将定位精度与网格分辨率解耦。大量实验验证了我们的范式：即使在最经济的3倍划分下，DISTA-Net++在CSO-mAP上超过7倍划分的最先进方法16.15%，在计数准确率上超过62.96%，而计算量仅为后者的六分之一，表明解混精度不必通过更细的离散化来换取。完整的生态系统可在以下网址获取：此https URL。
英文摘要
Long-range infrared imaging frequently confronts dense target clusters whose diffraction-limited signatures merge into a single indistinguishable blob, concealing the number, sub-pixel positions, and radiant intensities of the underlying sources. While deep learning has advanced general object detection, resolving such Closely-Spaced Infrared Small Targets (CSIST) remains largely unexplored, owing to a systemic infrastructure void and a fundamental paradigm mismatch. The dominant formulation, which reduces unmixing to a blind, discrete sub-pixel separation, is inherently insufficient: without semantic guidance, the ill-posed inverse problem admits ambiguous solutions plagued by false and missed detections, while grid-based discretization locks predictions onto fixed lattice centers, chaining precision to prohibitively expensive grid refinement. We argue that CSIST unmixing should instead be informed and continuous. To ground this paradigm shift, we establish the first comprehensive open-source ecosystem for the field, comprising the large-scale CSIST-100K benchmark, a tailored metric suite, and the GrokCSO toolkit. Upon this foundation, we propose DISTA-Net++, which anchors a dynamic deep unfolding backbone with two synergistic mechanisms: a Count-Guided Prior that injects the global target count as an explicit semantic constraint to regularize the solution space, and a Continuous Coordinate Rectification that regresses off-grid offsets to decouple localization accuracy from grid resolution. Extensive experiments validate our paradigm: even under the most economical 3x division, DISTA-Net++ surpasses 7x-division state-of-the-art methods by 16.15% in CSO-mAP and 62.96% in count accuracy at merely one-sixth of their computation, demonstrating that unmixing precision need not be purchased with finer discretization. The complete ecosystem is available at https://github.com/GrokCV/GrokDet.
