当前位置:网站首页>【MobileNet V3】《Searching for MobileNetV3》
【MobileNet V3】《Searching for MobileNetV3》
2022-07-02 06:26:00 【bryant_meng】


ICCV-2019
文章目录
1 Background and Motivation
【MobileNet】《MobileNets: Efficient Convolutional Neural Networks for Mobile Vision Applications》(CVPR-2017)
【MobileNet V2】《MobileNetV2:Inverted Residuals and Linear Bottlenecks》(CVPR-2018)
deliver the next generation of high accuracy efficient neural network models to power on-device computer vision
2 Related Work
手动设计网络
reducing the number of parameters -> reducing the number of operations (MAdds) -> reducing the actual measured latencyNAS
cell level -> block levelQuantization
knowledge distillation
3 Advantages / Contributions
NAS + 手动设计组装成 mobilenet v3 backbone,提出了 hard swish 激活函数(swish 改进版),提出了 Lite R-ASPP 分割头(R-ASPP 改进版),在分类、目标检测、分割数据集上速度和精度均有提升
4 Method
1)Network Search
Platform-Aware NAS for Blockwise Search(来自 MnastNet,稍微修改了一下 reward design 的权重)
NetAdapt for Layerwise Search
search per layer for the number of filters
maximizes △ A c c △ l a t e n c y \frac{\bigtriangleup Acc}{\bigtriangleup latency} △latency△Acc
2)Network Improvements
Redesigning Expensive Layers
search 后的网络头尾比较重,进行了优化
头部
channels 32 + ReLU or swish 减小到了 channels 16 + hard swish
尾部

Nonlinearities

s w i s h ( x ) = x ⋅ σ ( x ) swish(x) = x \cdot \sigma(x) swish(x)=x⋅σ(x)
swish activation function 虽然提升了网络精度,但对硬件部署不够友好,增加了计算时间,作者采取了如下的改进(piece-wise linear)
h − s w i s h ( x ) = x R e L U 6 ( x + 3 ) 6 h-swish(x) = x \frac{ReLU6(x+3)}{6} h−swish(x)=x6ReLU6(x+3)
比 relu 慢的
only use h-swish at the second half of the model(we find that most of the benefits swish are realized by using them only in the deeper layers)
Large squeeze-and-excite

v3 相比于 v2,采用了 SE 模块,SE 里面的 sigmoid 也是采用的 hard 形式,也即 R e L U 6 ( x + 3 ) 6 \frac{ReLU6(x+3)}{6} 6ReLU6(x+3)

作者把 SE 模块中的 squeeze fc 固定成 block 中 expand 通道数的 1/4(图 4 红√ 处)
no discernible latency cost
MobileNetV3 Definitions
5 Experiments
use single-threaded large core in all our measurements
5.1 Datasets
- ImageNet
- COCO
- Cityscapes
5.2 Classification


左上角最好




1)Impact of non-linearities
这里的 112 看的不是特别懂,N 越大按道理用的 h-swish 越多,速度要慢一些,怎么还快了
2)Impact of other components
5.3 Detection

mAP 离谱,哈哈
5.4 Segmentation
R-ASPP 基础上改进


6 Conclusion(own) / Future work
Pareto-optimal,帕累托最优(来自百度百科)
帕累托最优(Pareto Optimality),也称为帕累托效率(Pareto efficiency),是指资源分配的一种理想状态,假定固有的一群人和可分配的资源,从一种分配状态到另一种状态的变化中,在没有使任何人境况变坏的前提下,使得至少一个人变得更好,这就是帕累托改进或帕累托最优化。
帕累托最优状态就是不可能再有更多的帕累托改进的余地;换句话说,帕累托改进是达到帕累托最优的路径和方法。 帕累托最优是公平与效率的“理想王国”。是由帕累托提出的。MobileNet V3 = MobileNet v2 + SE + hard-swish activation + half initial layers channel & last block do global average pooling first(来自 盖肉特别慌)
边栏推荐
- Regular expressions in MySQL
- 【Mixup】《Mixup:Beyond Empirical Risk Minimization》
- conda常用命令
- [model distillation] tinybert: distilling Bert for natural language understanding
- Drawing mechanism of view (3)
- win10+vs2017+denseflow编译
- 【DIoU】《Distance-IoU Loss:Faster and Better Learning for Bounding Box Regression》
- 【Cascade FPD】《Deep Convolutional Network Cascade for Facial Point Detection》
- What if the notebook computer cannot run the CMD command
- ModuleNotFoundError: No module named ‘pytest‘
猜你喜欢

PointNet原理证明与理解

TimeCLR: A self-supervised contrastive learning framework for univariate time series representation

How to clean up logs on notebook computers to improve the response speed of web pages

【MagNet】《Progressive Semantic Segmentation》
![[introduction to information retrieval] Chapter 7 scoring calculation in search system](/img/cc/a5437cd36956e4c239889114b783c4.png)
[introduction to information retrieval] Chapter 7 scoring calculation in search system
![[model distillation] tinybert: distilling Bert for natural language understanding](/img/c1/e1c1a3cf039c4df1b59ef4b4afbcb2.png)
[model distillation] tinybert: distilling Bert for natural language understanding

生成模型与判别模型的区别与理解

【Programming】
![[paper introduction] r-drop: regulated dropout for neural networks](/img/09/4755e094b789b560c6b10323ebd5c1.png)
[paper introduction] r-drop: regulated dropout for neural networks

mmdetection训练自己的数据集--CVAT标注文件导出coco格式及相关操作
随机推荐
论文写作tip2
【AutoAugment】《AutoAugment:Learning Augmentation Policies from Data》
Label propagation
[CVPR‘22 Oral2] TAN: Temporal Alignment Networks for Long-term Video
Huawei machine test questions-20190417
Translation of the paper "written mathematical expression recognition with bidirectionally trained transformer"
[introduction to information retrieval] Chapter 3 fault tolerant retrieval
Regular expressions in MySQL
[torch] some ideas to solve the problem that the tensor parameters have gradients and the weight is not updated
PHP returns the corresponding key value according to the value in the two-dimensional array
PHP returns the abbreviation of the month according to the numerical month
【深度学习系列(八)】:Transoform原理及实战之原理篇
Record of problems in the construction process of IOD and detectron2
The difference and understanding between generative model and discriminant model
How to turn on night mode on laptop
Open failed: enoent (no such file or directory) / (operation not permitted)
超时停靠视频生成
[tricks] whiteningbert: an easy unsupervised sentence embedding approach
[paper introduction] r-drop: regulated dropout for neural networks
Jordan decomposition example of matrix


