作者：Chanho Kim，Li Fuxin，Mazen Alotaibi，James M. Rehg

作者单位：Georgia Institute of Technology，Oregon State University

发布时间：2021

发布期刊/会议：CVPR

出版商：IEEE

**论文全称：**Discriminative Appearance Modeling with Multi-track Pooling for Real-time Multi-object Tracking

论文地址：

CVPR 2021 Open Access Repository

论文代码：

https://github.com/chkim403/blstm-mtp

地位：

个人理解

• **为什么：**在多目标跟踪中，跟踪器在其内存中保存场景中每个目标的外观和运动信息。该内存中信息用于查找轨迹和检测框之间的匹配，并根据匹配结果对信息进行更新。但许多方法孤立地对每个目标建模，缺乏使用场景中所有目标共同更新记忆的能力。而这类方法在场景中出现具有相似外观的目标时，会导致跟踪效果不佳。
• 创新点：
  • 设计了一个新颖的多轨迹池化模块（multi-track pooling module）解决了在内存更新过程中同时考虑所有轨道的问题（使用正负样本来增强模型的外观辨别能力），并且空间开销很小
  • 提出了一种适合多轨迹池化的训练策略，该策略可以在线生成硬跟踪集
  • 使用双线性LSTM跟踪框架来构造强判别外观模型（Discriminative Appearance Model），在该框架下，可以使用贪婪数据关联来实现在线跟踪性能

一、摘要

In multi-object tracking, the tracker maintains in its memory the appearance and motion information for each object in the scene. This memory is utilized for finding matches between tracks and detections, and is updated based on the matching. Many approaches model each target in isolation and lack the ability to use all the targets in the scene to jointly update the memory. This can be problematic when there are similarly looking objects in the scene. In this paper, we solve the problem of simultaneously considering all tracks during memory updating, with only a small spatial overhead, via a novel multi-track pooling module. We additionally propose a training strategy adapted to multi-track pooling which generates hard tracking episodes online. We show that the combination of these innovations results in a strong discriminative appearance model under the bilinear LSTM tracking framework, enabling the use of greedy data association to achieve online tracking performance. Our experiments demonstrate real-time, state-of-the-art online tracking performance on public multi-object tracking (MOT) datasets.

• 设计了一个新颖的**多轨迹池化模块（multi-track pooling module）**解决了在内存更新过程中同时考虑所有轨道的问题，并且空间开销很小
• 提出了一种适合多轨迹池化的训练策略，该策略可以在线生成硬跟踪集
• 使用双线性LSTM跟踪框架来构造强判别外观模型（Discriminative Appearance Model），在该框架下，可以使用贪婪数据关联来实现在线跟踪性能