根据自车速度动态调整滤波器，提升高速场景下多目标追踪稳定性。
Stable at Any Speed: Speed-Driven Multi-Object Tracking with Learnable Kalman Filtering
- 基于自车速度自适应调节卡尔曼滤波不确定性参数，改进观测噪声建模。
- 在KITTI 2D MOT上达79.59% HOTA，nuScenes 3D MOT比SimpleTrack高2.2% AMOTA。
- 适合自动驾驶高动态场景下的多目标追踪，尤其对高速行驶时的稳定跟踪有帮助。
多目标追踪（MOT）使自动驾驶车辆能够持续感知动态物体，为预测、行为理解与安全规划提供关键的时间线索。然而，传统基于检测的追踪方法通常依赖于静态坐标变换，忽略自车速度引起的观测噪声变化和参考系偏移，导致在高速动态场景中追踪精度和稳定性下降。本文研究自车速度在MOT中的关键作用，提出速度引导的可学习卡尔曼滤波器（SG-LKF），能根据自车速度动态调整不确定性建模，显著提升复杂动态场景下的追踪性能。核心是运动尺度网络（MSNet），一种解耦的token混洗与通道混洗MLP，用于自适应预测SG-LKF的关键参数。为增强帧间关联与轨迹连续性，引入自监督轨迹一致性损失，联合优化语义与位置约束。大量实验表明，SG-LKF在KITTI 2D MOT上以79.59% HOTA排名第一，3D MOT上达82.03% HOTA，nuScenes 3D MOT上比SimpleTrack高2.2% AMOTA。
原文摘要 · Abstract (English)
Multi-object tracking (MOT) enables autonomous vehicles to continuously perceive dynamic objects, supplying essential temporal cues for prediction, behavior understanding, and safe planning. However, conventional tracking-by-detection methods typically rely on static coordinate transformations based on ego-vehicle poses, disregarding ego-vehicle speed-induced variations in observation noise and reference frame changes, which degrades tracking stability and accuracy in dynamic, high-speed scenarios. In this paper, we investigate the critical role of ego-vehicle speed in MOT and propose a Speed-Guided Learnable Kalman Filter (SG-LKF) that dynamically adapts uncertainty modeling to ego-vehicle speed, significantly improving stability and accuracy in highly dynamic scenarios. Central to SG-LKF is MotionScaleNet (MSNet), a decoupled token-mixing and channel-mixing MLP that adaptively predicts key parameters of SG-LKF. To enhance inter-frame association and trajectory continuity, we introduce a self-supervised trajectory consistency loss jointly optimized with semantic and positional constraints. Extensive experiments show that SG-LKF ranks first among all vision-based methods on KITTI 2D MOT with 79.59% HOTA, delivers strong results on KITTI 3D MOT with 82.03% HOTA, and outperforms SimpleTrack by 2.2% AMOTA on nuScenes 3D MOT.
Thank you to arXiv for use of its open access interoperability. PaperDance 不是 arXiv 官方产品；中文卡片由大模型生成，请以原文为准。