⚡️An Easy-to-use and Fast Deep Learning Model Deployment Toolkit for ☁️Cloud 📱Mobile and 📹Edge. Including Image, Video, Text and Audio 20+ main stream scenarios and 150+ SOTA models with end-to-end optimization, multi-platform and multi-framework support.
rainyfly
commented
7 months ago
是使用人脸识别项目的时候，手机前后摄像头，在手机移动的时候，会有上个画面帧残留，使用了小米手机和vivo手机，必现。 FastDeploy/blob/develop/examples/vision/facedet/scrfd/android