名称 发布 主要贡献
backbone
论文
deeplab v1 ICLR 2015
提出空洞卷积
引入条件随机
场
VGG16 Semantic Image Segmentation with
Deep Convolutional Nets and Fully
Connected CRFs
deeplab v2 CVPR 2016
引入空间金字
塔池化
（ ASPP ）
ResNet101 DeepLab: Semantic Image Segmentation
with Deep Convolutional Nets, Atrous
Convolution,and Fully Connected CRFs
deeplab v3 CVPR 2017
改进 ASPP
ResNet101 Rethinking Atrous Convolution for
Semantic Image Segmentation
deeplab v3+ ECCV 2018 encoder-
decoder 结构
改进后的
xception
Encoder-Decoder with Atrous Separable
Convolution for Semantic Image
Segmentation
