INT: Towards Infinite-frames 3D Detection with An Efficient Framework
It is natural to construct a multi-frame instead of a single-frame 3D detector for a continuous-time stream. Although increasing the number of frames might improve performance, previous multi-frame studies only used very limited frames to build their systems due to the dramatically increased computa...
Saved in:
Published in | arXiv.org |
---|---|
Main Authors | , , , , , , , , , |
Format | Paper |
Language | English |
Published |
Ithaca
Cornell University Library, arXiv.org
13.02.2023
|
Subjects | |
Online Access | Get full text |
Cover
Loading…
Abstract | It is natural to construct a multi-frame instead of a single-frame 3D detector for a continuous-time stream. Although increasing the number of frames might improve performance, previous multi-frame studies only used very limited frames to build their systems due to the dramatically increased computational and memory cost. To address these issues, we propose a novel on-stream training and prediction framework that, in theory, can employ an infinite number of frames while keeping the same amount of computation as a single-frame detector. This infinite framework (INT), which can be used with most existing detectors, is utilized, for example, on the popular CenterPoint, with significant latency reductions and performance improvements. We've also conducted extensive experiments on two large-scale datasets, nuScenes and Waymo Open Dataset, to demonstrate the scheme's effectiveness and efficiency. By employing INT on CenterPoint, we can get around 7% (Waymo) and 15% (nuScenes) performance boost with only 2~4ms latency overhead, and currently SOTA on the Waymo 3D Detection leaderboard. |
---|---|
AbstractList | It is natural to construct a multi-frame instead of a single-frame 3D detector for a continuous-time stream. Although increasing the number of frames might improve performance, previous multi-frame studies only used very limited frames to build their systems due to the dramatically increased computational and memory cost. To address these issues, we propose a novel on-stream training and prediction framework that, in theory, can employ an infinite number of frames while keeping the same amount of computation as a single-frame detector. This infinite framework (INT), which can be used with most existing detectors, is utilized, for example, on the popular CenterPoint, with significant latency reductions and performance improvements. We've also conducted extensive experiments on two large-scale datasets, nuScenes and Waymo Open Dataset, to demonstrate the scheme's effectiveness and efficiency. By employing INT on CenterPoint, we can get around 7% (Waymo) and 15% (nuScenes) performance boost with only 2~4ms latency overhead, and currently SOTA on the Waymo 3D Detection leaderboard. |
Author | Sun, Zhengyang Pan, Hongyu Zhang, Da Li, Hongmin Zhan, Xin Miao, Zhenwei Liu, Kaixuan Zhu, Jun Peihan Hao Xu, Jianyun |
Author_xml | – sequence: 1 givenname: Jianyun surname: Xu fullname: Xu, Jianyun – sequence: 2 givenname: Zhenwei surname: Miao fullname: Miao, Zhenwei – sequence: 3 givenname: Da surname: Zhang fullname: Zhang, Da – sequence: 4 givenname: Hongyu surname: Pan fullname: Pan, Hongyu – sequence: 5 givenname: Kaixuan surname: Liu fullname: Liu, Kaixuan – sequence: 6 fullname: Peihan Hao – sequence: 7 givenname: Jun surname: Zhu fullname: Zhu, Jun – sequence: 8 givenname: Zhengyang surname: Sun fullname: Sun, Zhengyang – sequence: 9 givenname: Hongmin surname: Li fullname: Li, Hongmin – sequence: 10 givenname: Xin surname: Zhan fullname: Zhan, Xin |
BookMark | eNqNykELgjAYgOERBVn5Hz7oLOiWU7pFKnnp5F3EvtGsvtU28e9X0A_o9B7eZ8XmZAhnLOBCJFG-43zJQueGOI65zHiaioCV9bnZQ2Omzl4c1KQ0aY-Rst0DHYgCCvTYe20IJu2vcCAoldK9RvJQfdVk7G3DFqq7Owx_XbNtVTbHU_S05jWi8-1gRkuf1fKMx1LKXCTiP_UGK387gA |
ContentType | Paper |
Copyright | 2023. This work is published under http://creativecommons.org/licenses/by-nc-nd/4.0/ (the “License”). Notwithstanding the ProQuest Terms and Conditions, you may use this content in accordance with the terms of the License. |
Copyright_xml | – notice: 2023. This work is published under http://creativecommons.org/licenses/by-nc-nd/4.0/ (the “License”). Notwithstanding the ProQuest Terms and Conditions, you may use this content in accordance with the terms of the License. |
DBID | 8FE 8FG ABJCF ABUWG AFKRA AZQEC BENPR BGLVJ CCPQU DWQXO HCIFZ L6V M7S PIMPY PQEST PQQKQ PQUKI PRINS PTHSS |
DatabaseName | ProQuest SciTech Collection ProQuest Technology Collection Materials Science & Engineering Collection ProQuest Central (Alumni) ProQuest Central ProQuest Central Essentials AUTh Library subscriptions: ProQuest Central Technology Collection ProQuest One Community College ProQuest Central SciTech Premium Collection (Proquest) (PQ_SDU_P3) ProQuest Engineering Collection Engineering Database Publicly Available Content (ProQuest) ProQuest One Academic Eastern Edition (DO NOT USE) ProQuest One Academic ProQuest One Academic UKI Edition ProQuest Central China Engineering Collection |
DatabaseTitle | Publicly Available Content Database Engineering Database Technology Collection ProQuest Central Essentials ProQuest One Academic Eastern Edition ProQuest Central (Alumni Edition) SciTech Premium Collection ProQuest One Community College ProQuest Technology Collection ProQuest SciTech Collection ProQuest Central China ProQuest Central ProQuest Engineering Collection ProQuest One Academic UKI Edition ProQuest Central Korea Materials Science & Engineering Collection ProQuest One Academic Engineering Collection |
DatabaseTitleList | Publicly Available Content Database |
Database_xml | – sequence: 1 dbid: 8FG name: ProQuest Technology Collection url: https://search.proquest.com/technologycollection1 sourceTypes: Aggregation Database |
DeliveryMethod | fulltext_linktorsrc |
Discipline | Physics |
EISSN | 2331-8422 |
Genre | Working Paper/Pre-Print |
GroupedDBID | 8FE 8FG ABJCF ABUWG AFKRA ALMA_UNASSIGNED_HOLDINGS AZQEC BENPR BGLVJ CCPQU DWQXO FRJ HCIFZ L6V M7S M~E PIMPY PQEST PQQKQ PQUKI PRINS PTHSS |
ID | FETCH-proquest_journals_27206668313 |
IEDL.DBID | 8FG |
IngestDate | Thu Oct 10 17:30:25 EDT 2024 |
IsOpenAccess | true |
IsPeerReviewed | false |
IsScholarly | false |
Language | English |
LinkModel | DirectLink |
MergedId | FETCHMERGED-proquest_journals_27206668313 |
OpenAccessLink | https://www.proquest.com/docview/2720666831?pq-origsite=%requestingapplication% |
PQID | 2720666831 |
PQPubID | 2050157 |
ParticipantIDs | proquest_journals_2720666831 |
PublicationCentury | 2000 |
PublicationDate | 20230213 |
PublicationDateYYYYMMDD | 2023-02-13 |
PublicationDate_xml | – month: 02 year: 2023 text: 20230213 day: 13 |
PublicationDecade | 2020 |
PublicationPlace | Ithaca |
PublicationPlace_xml | – name: Ithaca |
PublicationTitle | arXiv.org |
PublicationYear | 2023 |
Publisher | Cornell University Library, arXiv.org |
Publisher_xml | – name: Cornell University Library, arXiv.org |
SSID | ssj0002672553 |
Score | 3.4565558 |
SecondaryResourceType | preprint |
Snippet | It is natural to construct a multi-frame instead of a single-frame 3D detector for a continuous-time stream. Although increasing the number of frames might... |
SourceID | proquest |
SourceType | Aggregation Database |
SubjectTerms | Datasets Frames Performance enhancement |
Title | INT: Towards Infinite-frames 3D Detection with An Efficient Framework |
URI | https://www.proquest.com/docview/2720666831 |
hasFullText | 1 |
inHoldings | 1 |
isFullTextHit | |
isPrint | |
link | http://utb.summon.serialssolutions.com/2.0.0/link/0/eLvHCXMwfZ3NS8MwFMAfuiJ4m1-omyOg1yJJtrZ6EXWpm2AZUmG3kaQpeKlzrVf_dt8LrR6EHUMgISG8z1_eA7hyRawNlULU2ppwPEGHNSHWNboxwkalEdx42iKLZm_j5-Vk2Qbc6har7GSiF9TFh6UY-TXlC9HUTiS_W3-G1DWKsqttC41dCDhVwqOf4unTb4xFRDFazPKfmPW6I-1DsNBrtzmAHVcdwp5HLm19BGqe5bcs99hqzeZV-U7mX1gSLVUzOWVT13hOqmIULGX3FVO-3gOqCZZ2TNUxXKYqf5yF3d6r9nXUq7-zyBPooZvvToGh48Kt0ZGwGqWXiJPClHhh3FmHHqORZzDcttL59ukB7FOjdOKNuRxCr9l8uQtUp40Z-TsbQfCgssUrjl6-1Q8R638g |
link.rule.ids | 786,790,12792,21416,33406,33777,43633,43838 |
linkProvider | ProQuest |
linkToHtml | http://utb.summon.serialssolutions.com/2.0.0/link/0/eLvHCXMwfZ1LS8QwEIAH3UX05hMfqwb0WiRJN61eRNzWVtfiocLeSpKm4KWum_r_zYRWD8KeAwkZhnl-zABcmzqSCkchSqlVEE5dwhoj6ypuFdOiUYwqT1sUInsPnxfTRV9wsz1WOdhEb6jrT4018hvsF7pQO-b0fvkV4NYo7K72KzQ2YRxywVHP4_Tpt8bCROQiZv7PzHrfke7C-E0uzWoPNky7D1seudT2AJK8KO9I6bFVS_K2-cDwL2iQlrKEz8jMdJ6TagkWS8lDSxI_78G5CZIOTNUhXKVJ-ZgFw9tVrx22-vsLP4KRS_PNMRCXuFCtpGBaOuvForhWjRMYNdq4jFHxE5isu-l0_fElbGfl67ya58XLGezg0nRkjymfwKhbfZtz51o7deHl9wMm6X89 |
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%3Ajournal&rft.genre=article&rft.atitle=INT%3A+Towards+Infinite-frames+3D+Detection+with+An+Efficient+Framework&rft.jtitle=arXiv.org&rft.au=Xu%2C+Jianyun&rft.au=Miao%2C+Zhenwei&rft.au=Zhang%2C+Da&rft.au=Pan%2C+Hongyu&rft.date=2023-02-13&rft.pub=Cornell+University+Library%2C+arXiv.org&rft.eissn=2331-8422 |