【论文阅读】Visual place recognition: A survey from deep learning perspective

Visual place recognition: A survey from deep learning perspective

原文链接：PDF
[1] Zhang X , Wang L , Su Y . Visual Place Recognition: A Survey from Deep Learning Perspective[J]. Pattern Recognition, 2020.
公开数据集介绍及链接

个人觉得这篇文章介绍的非常全面，很适合初学者对于视觉位置识别（Visual Place Recognition, VPR）的深度学习方法的理解。
文章中介绍了很多中成熟的方法，从特征提取、匹配、多模态融合、多传感器等角度介绍，然后有表格介绍各方法和数据集。在这篇文章中介绍了许多公开数据集及其特点，并附上了数据集的链接。对于方法的介绍的话会有在哪种数据集下精度多少的描述。还是很值得细读的。

Abstract

Visual place recognition has attracted widespread research interest in multiple fields such as computer vision and robotics. Recently, researchers have employed advanced deep learning techniques to tackle this problem. While an increasing number of studies have proposed novel place recognition methods based on deep learning, few of them has provided a whole picture about how and to what extent deep learning has been utilized for this issue. In this paper, by delving into over 200 references, we present a comprehensive survey that covers various aspects of place recognition from deep learning perspective. We first present a brief introduction of deep learning and discuss its opportunities for recognizing places. After that, we focus on existing approaches built upon convolutional neural networks, including off-the-shelf and specifically designed models as well as novel image representations. We also discuss challenging problems in place recognition and present an extensive review of the corresponding datasets. To explore the future directions, we describe open issues and some new tools, for instance, generative adversarial networks, semantic scene understanding and multi-modality feature learning for this research topic. Finally, a conclusion is drawn for this paper.
视觉位置识别在计算机视觉和机器人等多个领域都有着广泛研究。近段时间，研究人员采用深度学习技术来解决这个问题。虽然越来越多的研究提出了基于深度学习的新颖位置识别方法，但很少有研究全面介绍深度学习怎样或在多大程度上被用于此问题。在本文中，通过深入研究 200 多篇参考文献，我们提出了一个全面的调研结果，涵盖基于深度学习的位置识别的各个方面。我们首先简要介绍深度学习并讨论其位置识别的应用。之后，我们专注于基于卷积神经网络的现有方法，包括现成的和专门设计的模型以及新颖的图像表示。我们还讨论了地点识别中具有挑战性的问题，并对相应的数据集进行了广泛的回顾。为了探索未来的方向，我们描述了开放性问题和一些新工具，例如，生成对抗网络、语义场景理解和多模态特征学习。最后，总结本文。

文章结构

1. Introduction
2. A brief introduction of deep learning
   2.1. What is deep learning?
   2.2. Convolutional neural network(CNN)
3. Visual place recognition: from the deep learning perspective
   3.1. Visual place recognition pipeline
   3.2. Visual place recognition and image retrieval
4. CNN-Based place recognition
   4.1. Using pre-trained CNN models
   4.2. Powerful image representations
   4.3. Using fine-tuned CNN models or new architectures
   4.4. Similarity measure
   4.5. Evaluation criteria
   4.6. Runtime performance
5. Place recognition datasets and challenging issues
   5.1. Generic and long-term datasets
   5.2. Specific datasets and challenging issues
6. New tools and open issues
   6.1. Beyond convolutional neural networks
   6.2. Semantic information
   6.3. Heterogeneous data
7. Conclusion and research directions

几个有用的图表

1. 视觉位置识别方法的时间发展线
   
2. 视觉位置识别流程图
   
3. CNN-based 方法比较
   
4. 描述符比较（分全局和局部两种）
   
5. 公开数据集一览表（原文中有附各数据集的地址）