Mobile Real-time Facial Expression Tracking with the Assistant of Public AI-as-a-Service

Public AI-as-a-Service (AIaaS) is a promising next-generation computing paradigm that attracts resource-limited mobile users to outsource their machine learning tasks. However, the time delay between cloud/edge servers and end users makes it hard for real-time mobile artificial intelligence applications. In this paper, we propose a real-time mobile facial expression tracking framework combining AIaaS and mobile local auxiliary computing, including facial expression tracking and the corresponding task offloading. Mobile facial expression tracking estimates the facial expression of the current frame with the help of sparse optical flow and neural network. Task offloading utilizes neural network to dynamically estimate the moment of task offloading. According to the experimental results in a real-world environment, the proposed framework successfully fulfills the mobile real-time facial expression tracking requirements with high accuracy. Moreover, compared to its benchmark, it effectively reduces the power consumption of mobile devices and the potential service cost for AIaaS.