[论文解读] WearMask: Fast In-browser Face Mask Detection with Serverless Edge Computing for COVID-19

Zekun Wang, Pengwei Wang+3 more|arXiv (Cornell University)|Jan 4, 2021

Face recognition and analysis参考文献 18被引用 27

一句话总结

WearMask 在任何网页浏览器中通过 serverless 边缘计算框架运行 YOLO-Fastest（通过 NCNN 和 WebAssembly）进行浏览器内人脸口罩检测，无需安装。

ABSTRACT

The COVID-19 epidemic has been a significant healthcare challenge in the United States. According to the Centers for Disease Control and Prevention (CDC), COVID-19 infection is transmitted predominately by respiratory droplets generated when people breathe, talk, cough, or sneeze. Wearing a mask is the primary, effective, and convenient method of blocking 80% of all respiratory infections. Therefore, many face mask detection and monitoring systems have been developed to provide effective supervision for hospitals, airports, publication transportation, sports venues, and retail locations. However, the current commercial face mask detection systems are typically bundled with specific software or hardware, impeding public accessibility. In this paper, we propose an in-browser serverless edge-computing based face mask detection solution, called Web-based efficient AI recognition of masks (WearMask), which can be deployed on any common devices (e.g., cell phones, tablets, computers) that have internet connections using web browsers, without installing any software. The serverless edge-computing design minimizes the extra hardware costs (e.g., specific devices or cloud computing servers). The contribution of the proposed method is to provide a holistic edge-computing framework of integrating (1) deep learning models (YOLO), (2) high-performance neural network inference computing framework (NCNN), and (3) a stack-based virtual machine (WebAssembly). For end-users, our web-based solution has advantages of (1) serverless edge-computing design with minimal device limitation and privacy risk, (2) installation free deployment, (3) low computing requirements, and (4) high detection speed. Our WearMask application has been launched with public access at facemask-detection.com.

研究动机与目标

• 通过降低安装门槛和硬件成本，推动便捷的口罩检测以帮助 COVID-19 期间的公共卫生。
• 提供一个与设备无关、无需安装的部署，便于在本地处理数据以保护用户隐私。
• 开发一个轻量、适用于无 GPU 的边缘设备的快速推理流水线。
• 展示将深度学习（YOLO）、高性能推理框架（NCNN）和 WebAssembly 集成以在浏览器中部署。

提出的方法

• 在混合真实口罩数据集（MAFA、WIDER FACE）及额外的互联网样本上训练一个轻量级的 YOLO-Fastest 检测器（EfficientNet-lite 编码器）。
• 将 PyTorch 模型转换为 NCNN 格式，并实现一个 C++ 推理流水线。
• 将推理流水线编译为 WebAssembly，在浏览器中通过 NCNN 以 WASM 模块运行。
• 实现一个无服务器边缘计算设计，在用户设备本地处理视频数据。
• 在 facemask-detection.com 上公开可访问的浏览器内演示，展示无需安装的部署。
• 对代表性边缘设备进行检测性能（mAP@0.5）和推理速度的基准测试。

实验结果

研究问题

• RQ1 在浏览器中、无服务器边缘计算是否能够在不需要安装或云资源的情况下实现准确且快速的口罩检测？
• RQ2 在典型边缘设备上运行 WearMask 时，检测准确度（mAP@0.5）和 FPS 可以达到多少？
• RQ3 WearMask 流水线如何整合 YOLO-Fastest、NCNN 与 WebAssembly 以实现设备无关部署？
• RQ4 在本地处理视频而非云端带来的隐私和部署方面有哪些好处？

主要发现

• YOLO-Fastest 模型在 IoU=0.5 时的平均精度（mAP@0.5）为 0.89，经过 120 个训练周期。
• WearMask 框架作为一个在浏览器中运行的应用，依托 NCNN 和 WebAssembly，实现无需安装、设备无关的边缘推理。
• 该系统通过在终端用户设备本地处理视频数据，强调低硬件要求和隐私，公开演示网站 facemask-detection.com。
• 该方法面向在公共设施和小型企业中实现实时场景的快速检测，提供比云端解决方案更具成本效益的替代方案。