MobileOne: An Improved One millisecond Mobile Backbone

Efficient neural network backbones for mobile devices are often optimized for metrics such as FLOPs or parameter count. However, these metrics may not correlate well with latency of the network when deployed on a mobile device. Therefore, we perform extensive analysis of different metrics by deployi...

Ausführliche Beschreibung

Gespeichert in:
Bibliographische Detailangaben
Veröffentlicht in:Proceedings (IEEE Computer Society Conference on Computer Vision and Pattern Recognition. Online) S. 7907 - 7917
Hauptverfasser: Vasu, Pavan Kumar Anasosalu, Gabriel, James, Zhu, Jeff, Tuzel, Oncel, Ranjan, Anurag
Format: Tagungsbericht
Sprache:Englisch
Veröffentlicht: IEEE 01.06.2023
Schlagworte:
ISSN:1063-6919
Online-Zugang:Volltext
Tags: Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
Abstract Efficient neural network backbones for mobile devices are often optimized for metrics such as FLOPs or parameter count. However, these metrics may not correlate well with latency of the network when deployed on a mobile device. Therefore, we perform extensive analysis of different metrics by deploying several mobile-friendly networks on a mobile device. We identify and analyze architectural and optimization bottlenecks in recent efficient neural networks and provide ways to mitigate these bottlenecks. To this end, we design an efficient backbone MobileOne, with variants achieving an inference time under 1 ms on an iPhone12 with 75.9% top-1 accuracy on ImageNet. We show that MobileOne achieves state-of-the-art performance within the efficient architectures while being many times faster on mobile. Our best model obtains similar performance on ImageNet as MobileFormer while being 38×faster. Our model obtains 2.3% better top-1 accuracy on ImageNet than EfficientNet at similar latency. Furthermore, we show that our model generalizes to multiple tasks - image classification, object detection, and semantic segmentation with significant improvements in latency and accuracy as compared to existing efficient architectures when deployed on a mobile device. Code and models are available at https://github.com/apple/ml-mobileone
AbstractList Efficient neural network backbones for mobile devices are often optimized for metrics such as FLOPs or parameter count. However, these metrics may not correlate well with latency of the network when deployed on a mobile device. Therefore, we perform extensive analysis of different metrics by deploying several mobile-friendly networks on a mobile device. We identify and analyze architectural and optimization bottlenecks in recent efficient neural networks and provide ways to mitigate these bottlenecks. To this end, we design an efficient backbone MobileOne, with variants achieving an inference time under 1 ms on an iPhone12 with 75.9% top-1 accuracy on ImageNet. We show that MobileOne achieves state-of-the-art performance within the efficient architectures while being many times faster on mobile. Our best model obtains similar performance on ImageNet as MobileFormer while being 38×faster. Our model obtains 2.3% better top-1 accuracy on ImageNet than EfficientNet at similar latency. Furthermore, we show that our model generalizes to multiple tasks - image classification, object detection, and semantic segmentation with significant improvements in latency and accuracy as compared to existing efficient architectures when deployed on a mobile device. Code and models are available at https://github.com/apple/ml-mobileone
Author Vasu, Pavan Kumar Anasosalu
Ranjan, Anurag
Zhu, Jeff
Gabriel, James
Tuzel, Oncel
Author_xml – sequence: 1
  givenname: Pavan Kumar Anasosalu
  surname: Vasu
  fullname: Vasu, Pavan Kumar Anasosalu
  organization: Apple
– sequence: 2
  givenname: James
  surname: Gabriel
  fullname: Gabriel, James
  organization: Apple
– sequence: 3
  givenname: Jeff
  surname: Zhu
  fullname: Zhu, Jeff
  organization: Apple
– sequence: 4
  givenname: Oncel
  surname: Tuzel
  fullname: Tuzel, Oncel
  organization: Apple
– sequence: 5
  givenname: Anurag
  surname: Ranjan
  fullname: Ranjan, Anurag
  organization: Apple
BookMark eNotjt1Kw0AQRldRsNa8QS_2BRJndje7Ge9qsFqoVES9LbvJBFbzUxIRfHsD9eqDw-HwXYuLfuhZiBVChgh0W368vObKKcoUKJ0BOGvOREKOCp2DBlRUnIsFgtWpJaQrkUzTJwBohWipWAj7PITY8r7nO7nu5bY7jsMP13IGsottGyeuhr6WJ03e--orzB9uxGXj24mT_12K983DW_mU7vaP23K9S6MC850qJPLsa29JQTAUDBbaWG3BsSnyJq8UsiMEdqpSFolDDbM9-w2EBvRSrE7dyMyH4xg7P_4eEOa60Vb_AUmrR70
CODEN IEEPAD
ContentType Conference Proceeding
DBID 6IE
6IH
CBEJK
RIE
RIO
DOI 10.1109/CVPR52729.2023.00764
DatabaseName IEEE Electronic Library (IEL) Conference Proceedings
IEEE Proceedings Order Plan (POP) 1998-present by volume
IEEE Xplore All Conference Proceedings
IEEE Electronic Library (IEL)
IEEE Proceedings Order Plans (POP) 1998-present
DatabaseTitleList
Database_xml – sequence: 1
  dbid: RIE
  name: IEEE Electronic Library (IEL)
  url: https://ieeexplore.ieee.org/
  sourceTypes: Publisher
DeliveryMethod fulltext_linktorsrc
Discipline Applied Sciences
EISBN 9798350301298
EISSN 1063-6919
EndPage 7917
ExternalDocumentID 10204436
Genre orig-research
GroupedDBID 6IE
6IH
6IL
6IN
AAWTH
ABLEC
ADZIZ
ALMA_UNASSIGNED_HOLDINGS
BEFXN
BFFAM
BGNUA
BKEBE
BPEOZ
CBEJK
CHZPO
IEGSK
IJVOP
OCL
RIE
RIL
RIO
ID FETCH-LOGICAL-i204t-2199aeada6920b49b4183463607e485f5c21e7910e72c2619ebd0da6a69f0bf03
IEDL.DBID RIE
ISICitedReferencesCount 204
ISICitedReferencesURI http://www.webofscience.com/api/gateway?GWVersion=2&SrcApp=Summon&SrcAuth=ProQuest&DestLinkType=CitingArticles&DestApp=WOS_CPL&KeyUT=001058542608027&url=https%3A%2F%2Fcvtisr.summon.serialssolutions.com%2F%23%21%2Fsearch%3Fho%3Df%26include.ft.matches%3Dt%26l%3Dnull%26q%3D
IngestDate Wed Aug 27 02:56:31 EDT 2025
IsPeerReviewed false
IsScholarly true
Language English
LinkModel DirectLink
MergedId FETCHMERGED-LOGICAL-i204t-2199aeada6920b49b4183463607e485f5c21e7910e72c2619ebd0da6a69f0bf03
PageCount 11
ParticipantIDs ieee_primary_10204436
PublicationCentury 2000
PublicationDate 2023-June
PublicationDateYYYYMMDD 2023-06-01
PublicationDate_xml – month: 06
  year: 2023
  text: 2023-June
PublicationDecade 2020
PublicationTitle Proceedings (IEEE Computer Society Conference on Computer Vision and Pattern Recognition. Online)
PublicationTitleAbbrev CVPR
PublicationYear 2023
Publisher IEEE
Publisher_xml – name: IEEE
SSID ssj0003211698
Score 2.6271367
Snippet Efficient neural network backbones for mobile devices are often optimized for metrics such as FLOPs or parameter count. However, these metrics may not...
SourceID ieee
SourceType Publisher
StartPage 7907
SubjectTerms Computational modeling
Computer architecture
Efficient and scalable vision
Mobile handsets
Neural networks
Object detection
Performance evaluation
Semantic segmentation
Title MobileOne: An Improved One millisecond Mobile Backbone
URI https://ieeexplore.ieee.org/document/10204436
WOSCitedRecordID wos001058542608027&url=https%3A%2F%2Fcvtisr.summon.serialssolutions.com%2F%23%21%2Fsearch%3Fho%3Df%26include.ft.matches%3Dt%26l%3Dnull%26q%3D
hasFullText 1
inHoldings 1
isFullTextHit
isPrint
link http://cvtisr.summon.serialssolutions.com/2.0.0/link/0/eLvHCXMwlV07T8MwED5BxcBUHkW85YHVxU2c2GaDiooBSoWg6lY59lmqkFLUB7-fcxIKCwNbZJ3s6JLzPezvPoAr8ikBg_HcJVZy6aXgOtcZV8Gg0yaTecWfMn5Uw6GeTMyoAatXWBhErC6fYTc-Vmf5fu7WsVRGFp4IKdN8G7aVUjVYa1NQSSmVyY1u4HE9Ya7749FLllD02I0c4bFJduwr8ItEpfIhg_Y_V9-Dzg8aj402fmYftrA8gHYTPrLGOJeHkD_NC7Lx5xJv2G3J6nIBSdAAi-RCs2VMfj2rxdidde_FvMQOvA3uX_sPvKFF4DN6gRWnPcZY-gFsbhJRSFNIMsuq75dCqbOQuaSHisIAVImLCRIWXpA0yQdRBJEeQauk6Y-BpU662H4l0yZIF3raWe9EMDSJt1rYE-hEPUw_6s4X028VnP4xfga7UdX1VapzaK0Wa7yAHfe5mi0Xl9X3-gLqBJSW
linkProvider IEEE
linkToHtml http://cvtisr.summon.serialssolutions.com/2.0.0/link/0/eLvHCXMwlV07T8MwED5BQYKpPIp444HVxUmcxGaDiqqItlSoVN0qx7GlqlKK-uD3c3ZCYWFgi6yTHV1yvof93Qdwiz7FGitzqkPFKc85oyIRMU2tNFrImCeeP2XUTft9MR7LQQVW91gYY4y_fGaa7tGf5edzvXalMrTwkHEeJduwE3MeBiVca1NSiTCZSaSoAHIBk3et0eAtDjF-bDqWcNcm23UW-EWj4r1Iu_7P9Q-g8YPHI4ONpzmELVMcQb0KIEllnstjSHrzDK38tTD35KEgZcEAJXCAOHqh6dKlvzkpxcij0rNsXpgGvLefhq0OrYgR6BRfYEVxl5EKfwGVyJBlXGYcDdN3_koNF7GNdRiYFAMBk4bapUgmyxlKo7xlmWXRCdQKnP4USKS5dg1YYiEt1zYQWuWaWYmT5EowdQYNp4fJR9n7YvKtgvM_xm9grzPsdSfd5_7LBew7tZcXqy6htlqszRXs6s_VdLm49t_uC9hml90
openUrl ctx_ver=Z39.88-2004&ctx_enc=info%3Aofi%2Fenc%3AUTF-8&rfr_id=info%3Asid%2Fsummon.serialssolutions.com&rft_val_fmt=info%3Aofi%2Ffmt%3Akev%3Amtx%3Abook&rft.genre=proceeding&rft.title=Proceedings+%28IEEE+Computer+Society+Conference+on+Computer+Vision+and+Pattern+Recognition.+Online%29&rft.atitle=MobileOne%3A+An+Improved+One+millisecond+Mobile+Backbone&rft.au=Vasu%2C+Pavan+Kumar+Anasosalu&rft.au=Gabriel%2C+James&rft.au=Zhu%2C+Jeff&rft.au=Tuzel%2C+Oncel&rft.date=2023-06-01&rft.pub=IEEE&rft.eissn=1063-6919&rft.spage=7907&rft.epage=7917&rft_id=info:doi/10.1109%2FCVPR52729.2023.00764&rft.externalDocID=10204436