Inverted Residuals and Linear Bottlenecks: Mobile Networks for Classification, Detection and Segmentation

计算机科学 残余物 瓶颈 人工智能 目标检测 分割 模式识别(心理学) 数据挖掘 机器学习 算法 嵌入式系统
作者
Andrew Howard,Andrey Zhmoginov,Liang-Chieh Chen,Mark Sandler,Menglong Zhu
出处
期刊:Cornell University - arXiv 被引量:473
摘要

In this paper we describe a new mobile architecture, MobileNetV2, that improves the state of the art performance of mobile models on multiple tasks and benchmarks as well as across a spectrum of different model sizes. We also describe efficient ways of applying these mobile models to object detection in a novel framework we call SSDLite. Additionally, we demonstrate how to build mobile semantic segmentation models through a reduced form of DeepLabv3 which we call Mobile DeepLabv3. The MobileNetV2 architecture is based on an inverted residual structure where the input and output of the residual block are thin bottleneck layers opposite to traditional residual models which use expanded representations in the input an MobileNetV2 uses lightweight depthwise convolutions to filter features in the intermediate expansion layer. Additionally, we find that it is important to remove non-linearities in the narrow layers in order to maintain representational power. We demonstrate that this improves performance and provide an intuition that led to this design. Finally, our approach allows decoupling of the input/output domains from the expressiveness of the transformation, which provides a convenient framework for further analysis. We measure our performance on Imagenet classification, COCO object detection, VOC image segmentation. We evaluate the trade-offs between accuracy, and number of operations measured by multiply-adds (MAdd), as well as the number of parameters

科研通智能强力驱动
Strongly Powered by AbleSci AI
更新
大幅提高文件上传限制,最高150M (2024-4-1)

科研通是完全免费的文献互助平台,具备全网最快的应助速度,最高的求助完成率。 对每一个文献求助,科研通都将尽心尽力,给求助人一个满意的交代。
实时播报
ocean应助西粤学采纳,获得20
3秒前
4秒前
orixero应助陶醉铁身采纳,获得30
4秒前
彳亍1117发布了新的文献求助10
7秒前
李健应助Rainstorm27采纳,获得10
7秒前
ybb发布了新的文献求助10
8秒前
8秒前
莉莉安发布了新的文献求助10
9秒前
licomen完成签到,获得积分20
9秒前
立青发布了新的文献求助10
9秒前
怕黑的道天完成签到 ,获得积分10
10秒前
liuxiaoyang完成签到,获得积分20
13秒前
Chenzhs发布了新的文献求助10
14秒前
怕孤单的念云完成签到,获得积分10
14秒前
jesuissi完成签到 ,获得积分10
14秒前
锦李发布了新的文献求助30
15秒前
17秒前
xyr完成签到,获得积分10
17秒前
发三篇SCI发布了新的文献求助10
18秒前
沅沅完成签到,获得积分10
19秒前
19秒前
21秒前
薰硝壤应助要没时间了采纳,获得30
21秒前
24秒前
缓慢醉卉发布了新的文献求助10
24秒前
欣喜大地完成签到 ,获得积分10
27秒前
科研通AI2S应助锦李采纳,获得10
28秒前
张朝程完成签到,获得积分10
28秒前
29秒前
思源应助难过的花生采纳,获得10
31秒前
研友_Z33EGZ发布了新的文献求助50
32秒前
35秒前
37秒前
37秒前
温暖的函完成签到 ,获得积分10
37秒前
40秒前
Y哦莫哦莫发布了新的文献求助20
40秒前
所所应助crystal采纳,获得10
43秒前
43秒前
fdd博发布了新的文献求助30
44秒前
高分求助中
Sustainability in Tides Chemistry 1500
TM 5-855-1(Fundamentals of protective design for conventional weapons) 1000
CLSI EP47 Evaluation of Reagent Carryover Effects on Test Results, 1st Edition 800
Threaded Harmony: A Sustainable Approach to Fashion 799
Livre et militantisme : La Cité éditeur 1958-1967 500
Retention of title in secured transactions law from a creditor's perspective: A comparative analysis of selected (non-)functional approaches 500
"Sixth plenary session of the Eighth Central Committee of the Communist Party of China" 400
热门求助领域 (近24小时)
化学 医学 生物 材料科学 工程类 有机化学 生物化学 物理 内科学 纳米技术 计算机科学 化学工程 复合材料 基因 遗传学 催化作用 物理化学 免疫学 量子力学 细胞生物学
热门帖子
关注 科研通微信公众号,转发送积分 3055393
求助须知:如何正确求助?哪些是违规求助? 2712170
关于积分的说明 7430007
捐赠科研通 2356998
什么是DOI,文献DOI怎么找? 1248385
科研通“疑难数据库(出版商)”最低求助积分说明 606700
版权声明 596093