论文标题
备忘录：带有内存的多目标跟踪
MeMOT: Multi-Object Tracking with Memory
论文作者
论文摘要
我们提出了一种在线跟踪算法，该算法在公共框架下执行对象检测和数据关联，能够在长时间后链接对象。通过保留大型时空存储器来存储追踪对象的身份嵌入，以及根据需要自适应引用和汇总有用的信息来实现这一点。我们的模型称为Memot，由三个主要模块组成，这些模块都是基于变压器的三个主要模块：1）假设生成，这些假设生成在当前的视频框架中产生对象建议； 2）记忆编码从每个跟踪对象的内存中提取核心信息的内存； 3）记忆解码可以同时解决对象检测和数据关联任务以进行多对象跟踪。当对广泛采用的MOT基准数据集进行评估时，Memot观察到非常具竞争力的性能。
We propose an online tracking algorithm that performs the object detection and data association under a common framework, capable of linking objects after a long time span. This is realized by preserving a large spatio-temporal memory to store the identity embeddings of the tracked objects, and by adaptively referencing and aggregating useful information from the memory as needed. Our model, called MeMOT, consists of three main modules that are all Transformer-based: 1) Hypothesis Generation that produce object proposals in the current video frame; 2) Memory Encoding that extracts the core information from the memory for each tracked object; and 3) Memory Decoding that solves the object detection and data association tasks simultaneously for multi-object tracking. When evaluated on widely adopted MOT benchmark datasets, MeMOT observes very competitive performance.
