<Macro-Micro Adversarial Network for Human Parsing> 摘要:在人体语义分割中,像素级别的分类损失在其低级局部不一致性和高级语义不一致性方面存在缺陷.对抗性网络的引入使用单个鉴别器来解决这两个问题.然而,两种类型的解析不一致是由不同的机制产生的,因此单个鉴别器很难解决它们.为解决这两种不一致问题,本文提出了宏观 - 微观对抗网络(MMAN).它有两个鉴别器,一个鉴别器Macro D作用于低分辨率标签图并且惩罚语义不一致性,例如错位的身体部位.另一
Consolidate multiple tasks or operations into a single computational unit. This pattern can increase compute resource utilization, and reduce the costs and management overhead associated with performing compute processing in cloud-hosted applications