xiaxzhou的博客

Learning Deconvolution Network for Semantic Segmentation首先介绍了基于FCN的语义分割，如论文：Fully Convolutional Networks for Semantic SegmentationSemantic Image Segmentation with Deep Convolutional Nets and Fully Co

Object Contour Detection with a Fully Convolutional Encoder-Decoder Network使用卷积编码解码网络检测主要目标的边缘网络结构是： 编码：VGG-16 解码：反池化-卷积-激活-dropout 卷积核:The number of channels of every decoder layer is properly d

Efficient Inference in Fully Connected CRFs with Gaussian Edge Potentials摘要：most state-of-the-art techniques for multi-class image segmentation and labeling use conditional random fields defined over p