Let there be Color!: Joint End-to-end Learning of Global and Local Image Priors for Automatic Image

着色原理 網絡結構爲: 輸入爲灰度圖像,輸出爲彩色圖像.網絡結構包括四個部分,Low-Level Feature Network用於提取圖像的低層特徵,提取低層特徵後,通過一個Mid-Level Feature Network得到中間特徵,另外通過Global Feature Network提取全局特徵,將全局特徵與中間特徵融合,即Fusion layer,之後將融合特徵輸入Colorizatio

>>阅读原文<<