第一次跳坑 POJ 3687 Labeling Balls Description Windy has N balls of distinct weights from 1 unit to N units. Now he tries to label them with 1 to N in such a way that: No two balls share the same label
Unsupervised Domain Adaptation via Structured Prediction Based Selective Pseudo-Labeling 基于结构化预测的选择性伪标记的无监督域自适应IntroductionRelated WorkProposed MethodExperiments and Results 基于结构化预测的选择性伪标记的无监
论文地址:High-Resolution Representations for Labeling Pixels and Regions 一篇使用HRNet进行语义分割和目标检测论文 HRNet=high resolution Net low resolution net就是用于分类的网络,通过stride逐步减少resolution,获取语义信息,最终得到分类 但是这种low-resolut