发表机构
Shanghai Jiao Tong University（上海交通大学）
机构由 AI 辅助整理，请以论文原文为准。AI 中文总结
该研究提出轻量型学习型稀疏特征跟踪器KLTNet，替换传统KLT跟踪器，结合由粗到精架构与各向同性置信度权重，在多数据集实验中提升了VIO系统的跟踪与里程计精度，且保持实时性。
AI 中文摘要
许多基于特征的视觉-惯性里程计（VIO）系统依赖于稀疏特征跟踪，其准确性和鲁棒性直接影响状态估计。传统KLT跟踪器主要依赖局部图像块，在快速运动或低纹理环境下可能变得不可靠。我们提出KLTNet，一种轻量型基于学习的即插即用稀疏特征跟踪器，旨在替换基于KLT的VIO前端中的传统KLT跟踪器。KLTNet遵循由粗到精、由密到疏的架构，结合低分辨率稠密光流实现鲁棒的全局运动初始化，并采用三元组块细化实现准确且时间一致的跟踪。固定参考块在每个特征跟踪过程中提供稳定锚点，有助于减少累积跟踪漂移。此外，KLTNet通过可微分多视图三角测量监督预测各向同性置信度权重，该权重可作为兼容VIO估计器中的观测权重。在公共基准数据集和自行收集的低纹理数据集上，结合VINS-Mono与OpenVINS的实验表明，相较于传统KLT，KLTNet提升了跟踪和里程计准确性，同时在嵌入式平台上保持实时性能。
英文摘要
Many feature-based visual-inertial odometry (VIO) systems rely on sparse feature tracking, whose accuracy and robustness directly affect state estimation. Classical KLT trackers rely primarily on local image patches and can become unreliable under rapid motion or in low-texture environments. We propose KLTNet, a lightweight learning-based, plug-and-play sparse feature tracker designed to replace classical KLT trackers in KLT-based VIO front ends. KLTNet follows a coarse-to-fine, dense-to-sparse architecture that combines low-resolution dense optical flow for robust global motion initialization with triplet-patch refinement for accurate and temporally consistent tracking. A fixed reference patch provides a stable anchor throughout each feature track and helps reduce accumulated tracking drift. In addition, KLTNet predicts anisotropic confidence weights supervised through differentiable multi-view triangulation, which can be used as observation weights in compatible VIO estimators. Experiments with VINS-Mono and OpenVINS on public benchmarks and a self-collected low-texture dataset demonstrate improved tracking and odometry accuracy over classical KLT, while maintaining real-time performance on an embedded platform.
