A 360-Degree Video Analytics Service for In-Classroom Firefighter Training
A 360-Degree Video Analytics Service for In-Classroom Firefighter Training
复制标题
DOI：
10.1109/cps-er56134.2022.00009
复制
发表时间：
2022-05
期刊：
影响因子：
--
通讯作者：
Ayush Sarkar;Anh Nguyen;Zhisheng Yan;K. Nahrstedt
中科院分区：
文献类型：
--
作者：
Ayush Sarkar;Anh Nguyen;Zhisheng Yan;K. Nahrstedt
通过视频演示搜索和救援任务中的消防行动是课堂消防员培训的常见方法。不幸的是，传统的2D摄像机有着根本性的弱点--它们只能捕捉到狭窄的视野，错过了大量来自消防员周围的信息，这在某些情况下可能会成为生死攸关的问题。本文提出了一种结合360°视频和深度学习优势的系统，可以自动检测全景场景中的重要对象，辅助消防教官进行课堂教学。具体来说，我们总结了显着的对象和事件相关的消防通过采访经验丰富的消防教练。利用这些知识，我们通过迁移学习方法研究了360°视频中消防物体的检测。我们报告了在通用对象和2D视频上训练的对象检测器的有见地的结果，并讨论了设计定制对象检测器的下一步。
Demonstrating firefighting operations in search and rescue missions through videos is a common approach to in-classroom firefighter training. Unfortunately, traditional 2D cameras have fundamental weaknesses – they can only capture a narrow field of view and miss a lot of information coming from the surroundings of the firefighter, which may become the matter of life and death in certain situations. In this paper, we propose a system combining the advantage of 360° videos and deep learning to automatically detect important objects in the panoramic scene, assisting firefighting instructors in classroom teaching scenarios. Specifically, we summarize the salient objects and events relevant to firefighting through an interview with an experienced firefighting instructor. Leveraging this knowledge, we investigate the detection of firefighting objects on 360° videos through a transfer learning approach. We report insightful results for object detectors trained on generic objects and 2D videos and discuss the next steps in designing a customized object detector.