FEELVOS：Fast End-to-End Embedding Learning for Video Object Segmentation

paper: https://arxiv.org/abs/1902.09513 code: https://github.com/kim-younghan/FEELVOS 1 Introduction 文章作者受《Blazingly Fast Video Object Segmentation with Pixel-Wise Metric Learning》的啓發，該方法在訓練過程中使用tripl

>>阅读原文<<