论文标题
转机：用于多个对象跟踪的全尺度改进方法
TransFiner: A Full-Scale Refinement Approach for Multiple Object Tracking
论文作者
论文摘要
多个对象跟踪（MOT）是包含检测和关联的任务。许多跟踪器已经取得了竞争性能。不幸的是，由于缺乏这些子任务的信息交流，在复杂的情况下，它们通常会偏向两者之一，并且在追踪人群中的个人时，不可避免的错过和错误的目标轨迹。本文提出了Transfiner，这是一种基于变压器的基于变压器的方法。这是一个通用的附件框架，取决于查询对，这是原始跟踪器和转工器之间的桥梁。通过融合解码器，每个查询对都会为特定对象产生精致的检测和运动线索。在此之前，它们在原始跟踪器的跟踪结果（位置和类预测）的指导下进行了特征和组标记，并以焦点和全面的方式完成跟踪改进。实验表明，在MOT17基准测试上，我们的设计是有效的，我们将中心拖车从67.8％的MOTA和64.7％的IDF1提升到71.5％MOTA和66.8％的IDF1。
Multiple object tracking (MOT) is the task containing detection and association. Plenty of trackers have achieved competitive performance. Unfortunately, for the lack of informative exchange on these subtasks, they are often biased toward one of the two and underperform in complex scenarios, such as the inevitable misses and mistaken trajectories of targets when tracking individuals within a crowd. This paper proposes TransFiner, a transformer-based approach to post-refining MOT. It is a generic attachment framework that depends on query pairs, the bridge between an original tracker and TransFiner. Each query pair, through the fusion decoder, produces refined detection and motion clues for a specific object. Before that, they are feature-aligned and group-labeled under the guidance of tracking results (locations and class predictions) from the original tracker, finishing tracking refinement with focus and comprehensively. Experiments show that our design is effective, on the MOT17 benchmark, we elevate the CenterTrack from 67.8% MOTA and 64.7% IDF1 to 71.5% MOTA and 66.8% IDF1.
