Global Tracking Transformers
Global Tracking Transformers
复制标题
DOI：
10.1109/cvpr52688.2022.00857
复制
发表时间：
2022-03
期刊：
影响因子：
--
通讯作者：
Xingyi Zhou;Tianwei Yin;V. Koltun;Philipp Krähenbühl
中科院分区：
文献类型：
--
作者：
Xingyi Zhou;Tianwei Yin;V. Koltun;Philipp Krähenbühl
我们提出了一种新的基于变换器的全局多目标跟踪架构。我们的网络将一个短的帧序列作为输入，并为所有对象生成全局轨迹。核心组件是一个全局跟踪Transformer，它对序列中所有帧中的对象进行操作。Transformer对来自所有帧的对象特征进行编码，并使用轨迹查询将它们分组为轨迹。轨迹查询是来自单个帧的对象特征，并且自然地产生独特的轨迹。我们的全局跟踪Transformer不需要中间成对分组或组合关联，并且可以与对象检测器联合训练。它在流行的MOT17基准上实现了具有竞争力的性能，MOTA为75.3，HOTA为59.1。更重要的是，我们的框架无缝集成到最先进的大词汇量检测器中，以跟踪任何对象。在具有挑战性的TAO数据集上的实验表明，我们的框架在基于成对关联的基线上不断改进，通过显着的7.7跟踪mAP超过了已发表的工作。代码可在https://github.com/xingyizhou/GTR上获得。
We present a novel transformer-based architecture for global multi-object tracking. Our network takes a short sequence of frames as input and produces global trajectories for all objects. The core component is a global tracking transformer that operates on objects from all frames in the sequence. The transformer encodes object features from all frames, and uses trajectory queries to group them into trajectories. The trajectory queries are object features from a single frame and naturally produce unique trajectories. Our global tracking transformer does not require intermediate pairwise grouping or combinatorial association, and can be jointly trained with an object detector. It achieves competitive performance on the popular MOT17 benchmark, with 75.3 MOTA and 59.1 HOTA. More importantly, our framework seamlessly integrates into state-of-the-art large-vocabulary detectors to track any objects. Experiments on the challenging TAO dataset show that our framework consistently improves upon baselines that are based on pairwise association, outperforming published work by a significant 7.7 tracking mAP. Code is available at https://github.com/xingyizhou/GTR.