Neural Enhanced Belief Propagation for Multiobject Tracking
Neural Enhanced Belief Propagation for Multiobject Tracking
复制标题
DOI：
10.1109/tsp.2023.3314275
复制
发表时间：
2022-12
影响因子：
5.4
通讯作者：
Mingchao Liang;Florian Meyer
中科院分区：
文献类型：
--
作者：
Mingchao Liang;Florian Meyer
多目标跟踪(MOT)的算法解决方案是自主导航和应用海洋科学应用的关键推动因素。最先进的MOT方法完全依赖于统计模型，通常使用经过预处理的传感器数据作为测量数据。具体地，由检测器产生测量结果，该检测器从以离散时间步长收集的原始传感器数据中提取潜在的目标位置。此准备处理步骤减少了数据流和计算复杂性，但可能会导致信息丢失。基于信任传播(BP)的最新贝叶斯MOT方法系统地利用统计模型的图结构来降低计算复杂性和提高可扩展性。然而，作为一种完全基于模型的方法，当统计模型和真实数据生成过程之间存在不匹配时，BP可以提供高度次优的估计。现有的基于BP的MOT方法只能进一步利用经过预处理的测量数据。在本文中，我们介绍了一种结合了基于模型和数据驱动的MOT的BP算法。提出的神经增强型信念传播(NEBP)方法通过从原始传感器数据中学习信息来补充BP的统计模型。该方法推测，学习的信息可以减少模型失配，从而改善数据关联和虚警抑制。与基于模型的方法相比，我们的NEBP方法提高了跟踪性能。同时，它继承了基于BP的MOT的优点，即只在目标数量上进行二次缩放，从而可以生成和维护大量的目标轨迹。我们在nuScenes自动驾驶数据集上对我们的NEBP方法进行了性能评估，并证明了它可以达到最先进的性能。特别是，多目标跟踪的平均准确率达到0.683，与非BP方法相比，身份开关和轨迹碎片分别降低了23%和19%。
Algorithmic solutions for multi-object tracking (MOT) are a key enabler for applications in autonomous navigation and applied ocean sciences. State-of-the-art MOT methods fully rely on a statistical model and typically use preprocessed sensor data as measurements. In particular, measurements are produced by a detector that extracts potential object locations from the raw sensor data collected at discrete time steps. This preparatory processing step reduces data flow and computational complexity but may result in a loss of information. State-of-the-art Bayesian MOT methods that are based on belief propagation (BP) systematically exploit graph structures of the statistical model to reduce computational complexity and improve scalability. However, as a fully model-based approach, BP can provide highly suboptimal estimates when there is a mismatch between the statistical model and the true data-generating process. Existing BP-based MOT methods can further only make use of preprocessed measurements. In this paper, we introduce a variant of BP that combines model-based with data-driven MOT. The proposed neural enhanced belief propagation (NEBP) method complements the statistical model of BP by information learned from raw sensor data. This approach conjectures that the learned information can reduce model mismatch and thus improve data association and false alarm rejection. Our NEBP method improves tracking performance compared to model-based methods. At the same time, it inherits the advantages of BP-based MOT, i.e., it scales only quadratically in the number of objects, and it can thus generate and maintain a large number of object tracks. We evaluate the performance of our NEBP approach for MOT on the nuScenes autonomous driving dataset and demonstrate that it can achieve state-of-the-art performance. In particular, an average multi-object tracking accuracy (AMOTA) of 0.683 was obtained and, compared with non-BP-based methods, identity switches (IDS) and track fragments (Frag) were reduced by 23% and 19%, respectively.