[论文解读] Deep Learning for Person Re-identification: A Survey and Outlook

Mang Ye, Jianbing Shen+4 more|arXiv (Cornell University)|Jan 13, 2020

Video Surveillance and Tracking Methods被引用 120

一句话总结

本综述分析了封闭世界与开放世界的人体再识别（Re-ID），提出了 AGW 基线和新的 mINP 指标，并概述了开放挑战与未来方向。

ABSTRACT

Person re-identification (Re-ID) aims at retrieving a person of interest across multiple non-overlapping cameras. With the advancement of deep neural networks and increasing demand of intelligent video surveillance, it has gained significantly increased interest in the computer vision community. By dissecting the involved components in developing a person Re-ID system, we categorize it into the closed-world and open-world settings. The widely studied closed-world setting is usually applied under various research-oriented assumptions, and has achieved inspiring success using deep learning techniques on a number of datasets. We first conduct a comprehensive overview with in-depth analysis for closed-world person Re-ID from three different perspectives, including deep feature representation learning, deep metric learning and ranking optimization. With the performance saturation under closed-world setting, the research focus for person Re-ID has recently shifted to the open-world setting, facing more challenging issues. This setting is closer to practical applications under specific scenarios. We summarize the open-world Re-ID in terms of five different aspects. By analyzing the advantages of existing methods, we design a powerful AGW baseline, achieving state-of-the-art or at least comparable performance on twelve datasets for FOUR different Re-ID tasks. Meanwhile, we introduce a new evaluation metric (mINP) for person Re-ID, indicating the cost for finding all the correct matches, which provides an additional criteria to evaluate the Re-ID system for real applications. Finally, some important yet under-investigated open issues are discussed.

研究动机与目标

• 总结基于深度学习的封闭世界人体再识别的主要组成部分与挑战。
• 区分封闭世界与开放世界的 Re-ID 并分析各自的方法学。
• 提出一个强基线（AGW）和一个新的评估指标（mINP），以推动未来研究。
• 讨论现实世界 Re-ID 系统的开放性问题与实际考虑。

提出的方法

• 回顾封闭世界 Re-ID 中的深度特征表示学习、深度度量学习和排序优化。
• 介绍并分析四种特征学习策略：全局特征、局部特征、辅助特征和视频特征。
• 讨论损失函数（识别、验证、三元组、OIM）及训练深度 Re-ID 模型的采样策略。
• 提出 AGW（带权三元组损失的注意力广义均值池化）作为一个强大的基线。
• 引入 mINP（平均逆负惩罚）作为辅助评估指标。

实验结果

研究问题

• RQ1 当前基于深度学习的封闭世界人体再识别方法的优点与局限性是什么？
• RQ2 开放世界设置如何带来额外挑战，哪些方法能够应对它们？
• RQ3 跨数据集和任务的强大、具有通用性的 Re-ID 基线应具备哪些特征？
• RQ4 新的 mINP 指标如何补充现有的 CMC/mAP 来评估 Re-ID 系统？

主要发现

• AGW 基线在四个 Re-ID 任务的十二个数据集上达到最先进或有竞争力的性能。
• mINP 提供了额外的评判标准，通过衡量找到所有正确匹配所需的成本。
• 开放世界挑战包括异质数据、端到端检索、有限/有噪声标签、标签噪声，以及开放集检索。
• 确定的开放问题旨在弥合封闭世界基准与现实世界部署之间的差距。