Tracking a current and/or previous position, velocity, acceleration, and/or heading of an object using sensor data may include determining whether to associate a current object detection generated from recently received (e.g., current) sensor data with a previous object detection generated from previously received sensor data. In other words, tracking may identify that an object detected in previous sensor data is the same as an object detected in current sensor data. However, multiple types of sensor data may be used to detect objects, and some objects may not be detected by different sensor types or may be detected in different ways, which may obfuscate attempts to track objects. An ML model may be trained to receive outputs associated with different sensor types and/or tracks associated with an object and determine a data structure including a region of interest, an object classification, and/or a pose associated with the object.
使用传感器数据跟踪对象的当前和/或先前位置、速度、加速度和/或航向可包括确定是否将根据最近接收的(例如,当前)传感器数据生成的当前对象检测与从以前接收的传感器数据生成的先前对象检测相关联。换言之,跟踪可以识别在以前的传感器数据中检测到的对象与在当前传感器数据中检测到的对象相同。然而,可以使用多种类型的传感器数据来检测对象,并且一些对象可能无法被不同的传感器类型检测到或者可能以不同的方式被检测到,这可能会混淆跟踪对象的尝试。可以训练ML模型以接收与不同传感器类型相关联的输出和/或与对象相关联的跟踪,并确定数据结构,该数据结构包括感兴趣区域、对象分类和/或与对象相关联的姿态。
Object detection and tracking
对象检测以及跟踪
2022-08-12
Patent
Elektronische Ressource
Chinesisch
Object Detection, Classification, and Tracking
Springer Verlag | 2017
|OBJECT TRACKING SYSTEM, OBJECT TRACKING METHOD, AND OBJECT TRACKING PROGRAM
Europäisches Patentamt | 2021
|