发表机构
Faculty of Information Technology, City University, Malaysia; Kyung Hee University; National University of Sciences and Technology, Pakistan; School of Computing and Artificial Intelligence, Sunway University, Malaysia; King Fahd University of Petroleum and Minerals, Saudi Arabia; SDAIA-KFUPM Joint Research Center for Artificial Intelligence, Saudi Arabia（马来西亚城市大学信息技术学院; 庆熙大学; 巴基斯坦国家科技大学; 马来西亚双威大学计算与人工智能学院; 沙特阿拉伯法赫德国王石油和矿产大学; 沙特阿拉伯SDAIA-KFUPM人工智能联合研究中心）
机构由 AI 辅助整理，请以论文原文为准。AI 中文总结
本文提出ROI门控SAHI框架，通过轻量级提议器限制切片式细化范围，在稀疏场景中可实现最高6.90倍加速，结合自适应路由策略能平衡速度与检测性能。
AI 中文摘要
切片辅助超推理（SAHI）可提升高分辨率图像中的小目标检测性能，但常需在背景切片上消耗大量计算资源。我们提出区域-of-interest（ROI）门控SAHI，这是一种推理时框架，引入轻量级提议器以定位前景区域，并将切片式细化限制在信息丰富的区域。我们在两种设置中评估该框架：在包含128张图像的COCO128完整划分数据集上，静态ROI门控平均速度慢于全SAHI，速度比为0.88，且mAP@0.5为0.6602，低于全SAHI的0.7569；采用τ=0.4的简单自适应路由策略可降低平均延迟，实现比全SAHI提升1.02倍的性能。在三张图像的稀疏到密集案例研究中，ROI门控实现0.96倍至6.90倍的加速，平均加速为3.41倍。这些结果表明，ROI门控在稀疏场景中最具优势，且需要基于策略的路由以实现稳健的平均性能。
英文摘要
Slicing-Aided Hyper Inference (SAHI) improves small object detection in high-resolution images but often spends substantial compute on background tiles. We propose region-of-interest (ROI)-Gated SAHI, an inference-time framework that introduces a lightweight proposer to localize foreground regions and restrict sliced refinement to informative areas. We evaluate the framework in two settings. On the COCO128 full split dataset comprising 128 images, static ROI-gating is slower on average than Full SAHI, achieving a speed ratio of 0.88, and yields a lower mAP@0.5 of 0.6602 compared with 0.7569 for Full SAHI. A simple adaptive routing policy with $τ=$ 0.4 educes the mean latency, achieving a slight gain of 1.02$\times$ over Full SAHI. On a three-image sparse-to-dense case study, ROI-gating achieves speedups ranging from 0.96$\times$ to 6.90$\times$ with a mean speedup of 3.41$\times$. These results show that ROI-gating is most beneficial in sparse scenes and requires policy-based routing for robust average behavior.
