⚡️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.
为什么会有这样的需求呢, v3的rec模型效果比v2是好的. 如果真想这么做的话,可以使用PP-OCRv2的demo用例,直接把v2的det模型 换成v3的det模型即可