当前位置: X-MOL 学术Image Vis. Comput. › 论文详情
Our official English website, www.x-mol.net, welcomes your feedback! (Note: you will need to create a separate account there.)
Person search: New paradigm of person re-identification: A survey and outlook of recent works
Image and Vision Computing ( IF 4.7 ) Pub Date : 2020-06-30 , DOI: 10.1016/j.imavis.2020.103970
Khawar Islam

Person Search (PS) has become a major field because of its need in community and in the field of research among researchers. This task aims to find a probe person from whole scene which shows great significance in video surveillance field to track lost people, re-identification, and verification of person. In last few years, deep learning has played unremarkable role for the solution of re-identification problem. Deep learning shows incredible performance in person (re-ID) and search. Researchers experience more flexibility in proposing new methods and solve challenging issues such as low resolution, pose variation, background clutter, occlusion, viewpoints, and low illumination. Specially, convolutional neural network (CNN) achieves breakthrough performance and extracts useful patterns and characteristics. Development of new framework takes substantial efforts; hard work and computation cost are required to acquire excellent results. This survey paper includes brief discussion about feature representation learning and deep metric learning with novel loss functions. We thoroughly review datasets with performance analysis on existing datasets. Finally, we are reviewing current solutions for further consideration.



中文翻译:

人物搜寻:人物重新识别的新范式:近期作品的调查与展望

人员搜索(PS)已成为一个主要领域,因为它在社区和研究人员之间的研究领域中均具有需求。这项任务旨在从整个场景中找到一个被调查人员,这在视频监视领域中对丢失人员进行跟踪,重新识别和验证人员具有重要意义。在过去的几年中,深度学习对于解决重新识别问题起着不显着的作用。深度学习显示了亲自(re-ID)和搜索的出色表现。研究人员在提出新方法和解决诸如低分辨率,姿势变化,背景杂波,遮挡,视点和低照度之类的挑战性问题时具有更大的灵活性。特别地,卷积神经网络(CNN)取得了突破性的性能,并提取了有用的模式和特征。新框架的开发需要大量的努力;获得出色的结果需要付出艰辛的工作和计算成本。本调查论文简要讨论了具有新损失函数的特征表示学习和深度度量学习。我们通过对现有数据集进行性能分析来彻底审查数据集。最后,我们正在审查当前的解决方案,以供进一步考虑。

更新日期:2020-06-30
down
wechat
bug