发表机构
University of Zaragoza; University of Palermo; The Chinese University of Hong Kong; Tohoku University; Central South University; Texas A&M University; Nanjing University of Aeronautics and Astronautics; The Hong Kong University of Science and Technology (Guangzhou)（萨拉戈萨大学; 巴勒莫大学; 香港中文大学; 东北大学; 中南大学; 德克萨斯农工大学; 南京航空航天大学; 香港科技大学（广州））
机构由 AI 辅助整理，请以论文原文为准。AI 中文总结
针对运动模糊下关键点检测的挑战，提出无去模糊的自监督关键点检测器SSMB，通过LDE模块及两阶段训练实现，在多项任务上优于现有基线，达到稀疏关键点检测的新最优性能。
AI 中文摘要
运动模糊下的关键点检测仍是一项重大挑战，因为模糊会扭曲图像局部结构并降低特征定位的可重复性。现有方法要么依赖计算成本高昂的“先去模糊再检测”流水线，可能引入恢复伪影；要么学习回归清晰图像上手工设计关键点的图像位置，这反映了手工检测器的假设，而非模糊下真正可重复的特性。我们提出SSMB，一种针对运动模糊图像的无去模糊、自监督关键点检测器，既不需要手工设计检测器，也不需要外部伪标签。SSMB引入局部可区分性增强（Local Discriminability Enhancement，LDE）模块，用于在全局特征混合后恢复细粒度的局部可区分性。训练分两个阶段进行：第一阶段，在合成形状上进行几何预训练，无需任何外部检测器，仅通过渲染几何结构即可启动空间可区分关键点检测；第二阶段，在真实清晰-模糊图像对上进行感知模糊的训练，通过多组件自监督目标学习模糊不变检测，该目标需满足跨域一致性、几何对齐和空间覆盖性。对运动模糊下的关键点检测、图像匹配、相对位姿估计和视觉定位进行的大量评估表明，SSMB在稀疏关键点检测器中达到了新的 state-of-the-art，在所有任务上始终优于监督和自监督基线。代码、模型和数据集将在论文接收后公开。
英文摘要
Keypoint detection under motion blur remains a significant challenge, as blur distorts local image structure and degrades the repeatability of feature localization. Existing approaches either rely on computationally expensive deblur-then-detect pipelines that may introduce restoration artifacts, or learn to regress the image positions of handcrafted keypoints extracted on sharp images, which reflects the assumptions of the handcrafted detector rather than what is truly repeatable under blur. We present SSMB, a deblur-free, self-supervised keypoint detector for motion-blurred images that requires neither handcrafted detectors nor external pseudo-labels. SSMB introduces the Local Discriminability Enhancement (LDE) module, which restores fine-grained local discriminability after global feature mixing. Training is performed in two stages. First, geometric pretraining on synthetic shapes bootstraps spatially discriminative keypoint detection without any external detector, just from the rendered geometry. Second, blur-aware training on real sharp-blur image pairs learns blur-invariant detection through a multi-component self-supervised objective that enforces cross-domain consistency, geometric alignment, and spatial coverage. Extensive evaluations on keypoint detection, image matching, relative pose estimation, and visual localization under motion blur demonstrate that SSMB establishes a new state-of-the-art among sparse keypoint detectors, consistently outperforming both supervised and self-supervised baselines across all tasks. Code, models, and datasets will be publicly available upon paper acceptance.
Comments20 pages, 11 figures, 14 tables
