DeepLab：基于深度卷积网络、宏卷积和全连接CRF的语义图像分割
提出空洞卷积（解决分辨率损失）和 CRF 后处理（优化边界），大幅提升分割精度，是遥感分割中处理多尺度地物的重要参考
Abstract
本论文使用深度学习来进行语义分割的任务，有三大贡献：
1、使用空洞卷积代替传统的卷积（使用上采样滤波器的卷积或“atrous卷积”，作为密集预测任务中的强大工具）
暂无评论
发送评论 编辑评论
正在回复 的评论 :
推荐文章
DDRL: Domain Distribution Reconstruction Learning for Binary Change Detection in Remote Sensing Images
Remote Sensing Change Detection With Transformers Trained From Scratch
A Multitask CNN-Transformer Network for Semantic Change Detection From Bitemporal Remote Sensing Images
VcT: Visual Change Transformer for Remote Sensing Image Change Detection
遥感图像变化检测
Multimodal Visual-Language Prompt Network for Remote Sensing Few-Shot Segmentation
MCKTNet: Multiscale Cross-Modal Knowledge Transfer Network for Semantic Segmentation of Remote Sensing Images
CFFormer: A Cross-Fusion Transformer Framework for the Semantic Segmentation of Multisource Remote Sensing Images
MetaSegNet: Metadata-Collaborative Vision-Language Representation Learning for Semantic Segmentation of Remote Sensing Images
Interactive Concept Network Enhanced Transformer for Remote Sensing Image Captioning