Paper reading: ISDNet: Integrating Shallow and Deep Networks CVPR2022

ISDNet: Integrating Shallow and Deep Networks for Efficient Ultra-high Resolution Segmentation CVPR2022

论文介绍:在高分辨率的图像分割中,计算和内存的需求都是非常大的。原有的方法主要是采用全局-局部细化的路线,虽然能够很好地考虑到内存的消耗,但是忽略了推断速度。本文主要关注于在整张图像上进行直接推理。本文提出一个高分辨率图像分割框架(Integrating Shallow and Deep Networks, ISDNet),他能很好地整合浅层和深层网络,并显著的提高推理速度并有一个精准的分割结果。为了更好地利用浅层特征和深层特征之间的关联性,提出一个关系感知特征融合模块(Relational-Aware feature Fusion module),以保证网络分割的性能和鲁棒性。

高分辨率图像分割框架对比图

本文的分割网络框架图

深层网络和浅层网络的输入和输出

High-frequency residuals的计算方式如下: 

High-frequency residuals处理之后的图像,作为浅层网络的输入,以学习那些互补的空间细节。对于深层网络,是把原始图像下采样为小尺度图像,作为输入。深层网络分支有三个损失,即:

辅助分割头(an auxiliary segmentation head) :标准的cross-entropy函数

超分辨率头(a super-resolution head)

结构蒸馏损失(a structure distillation loss):

 

最后浅层网络输出1/8和1/16的特征图;深层网络输出1/32的特征图。

特征融合部分:(浅层网络+深层网络特征的关系感知特征融合模块)

 

首先对深层特征和浅层特征图进行channel-wise attention操作,之后通过內积,求解两者的关联矩阵,最后融合:

实验分析

 (1)DeepGlobe. The DeepGlobe dataset contains 803 images with 2448 × 2448 resolution. It contains 7 classes of landscape regions, in which the class named ”unknown” is not considered in the evaluation. We follow the protocol as [3], by splitting images all of the images into training,
validation and test set with 455, 207 and 142 images respectively.

(2)Inria Aerial. The Inria Aerial dataset provides 180 images with 5000 × 5000 resolution and dense annotations with a binary mask for building and non-building areas. Following [3], we split images into training, validation and test set with 126, 27 and 27 images respectively.

(3)Cityscapes. The Cityscapes dataset is a popular generic dataset for semantic segmentation, which has 5,000 fine annotated images with 1024×2048 resolution. We follow the
official data split for our experiments, which contains 2,975 images for training, 500 images for validation and the rest 1525 images for testing.

 

 

 

评论
添加红包

请填写红包祝福语或标题

红包个数最小为10个

红包金额最低5元

当前余额3.43前往充值 >
需支付:10.00
成就一亿技术人!
领取后你会自动成为博主和红包主的粉丝 规则
hope_wisdom
发出的红包
实付
使用余额支付
点击重新获取
扫码支付
钱包余额 0

抵扣说明:

1.余额是钱包充值的虚拟货币,按照1:1的比例进行支付金额的抵扣。
2.余额无法直接购买下载,可以购买VIP、付费专栏及课程。

余额充值