Divide-and-Conquer for Lane-Aware Diverse Trajectory Prediction

Trajectory prediction is a safety-critical tool for autonomous vehicles to plan and execute actions. Our work addresses two key challenges in trajectory prediction, learning multimodal outputs, and better predictions by imposing constraints using driving knowledge. Recent methods have achieved stron...

Full description

Saved in:
Bibliographic Details
Published in2021 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR) pp. 15794 - 15803
Main Authors Narayanan, Sriram, Moslemi, Ramin, Pittaluga, Francesco, Liu, Buyu, Chandraker, Manmohan
Format Conference Proceeding
LanguageEnglish
Published IEEE 01.01.2021
Subjects
Online AccessGet full text

Cover

Loading…
Abstract Trajectory prediction is a safety-critical tool for autonomous vehicles to plan and execute actions. Our work addresses two key challenges in trajectory prediction, learning multimodal outputs, and better predictions by imposing constraints using driving knowledge. Recent methods have achieved strong performances using Multi-Choice Learning objectives like winner-takes-all (WTA) or best-of-many. But the impact of those methods in learning diverse hypotheses is under-studied as such objectives highly depend on their initialization for diversity. As our first contribution, we propose a novel Divide-And-Conquer (DAC) approach that acts as a better initialization technique to WTA objective, resulting in diverse outputs without any spurious modes. Our second contribution is a novel trajectory prediction framework called ALAN that uses existing lane centerlines as anchors to provide trajectories constrained to the input lanes. Our framework provides multi-agent trajectory outputs in a forward pass by capturing interactions through hypercolumn descriptors and incorporating scene information in the form of rasterized images and per-agent lane anchors. Experiments on synthetic and real data show that the proposed DAC captures the data distribution better compare to other WTA family of objectives. Further, we show that our ALAN approach provides on par or better performance with SOTA methods evaluated on Nuscenes urban driving benchmark.
AbstractList Trajectory prediction is a safety-critical tool for autonomous vehicles to plan and execute actions. Our work addresses two key challenges in trajectory prediction, learning multimodal outputs, and better predictions by imposing constraints using driving knowledge. Recent methods have achieved strong performances using Multi-Choice Learning objectives like winner-takes-all (WTA) or best-of-many. But the impact of those methods in learning diverse hypotheses is under-studied as such objectives highly depend on their initialization for diversity. As our first contribution, we propose a novel Divide-And-Conquer (DAC) approach that acts as a better initialization technique to WTA objective, resulting in diverse outputs without any spurious modes. Our second contribution is a novel trajectory prediction framework called ALAN that uses existing lane centerlines as anchors to provide trajectories constrained to the input lanes. Our framework provides multi-agent trajectory outputs in a forward pass by capturing interactions through hypercolumn descriptors and incorporating scene information in the form of rasterized images and per-agent lane anchors. Experiments on synthetic and real data show that the proposed DAC captures the data distribution better compare to other WTA family of objectives. Further, we show that our ALAN approach provides on par or better performance with SOTA methods evaluated on Nuscenes urban driving benchmark.
Author Narayanan, Sriram
Pittaluga, Francesco
Chandraker, Manmohan
Moslemi, Ramin
Liu, Buyu
Author_xml – sequence: 1
  givenname: Sriram
  surname: Narayanan
  fullname: Narayanan, Sriram
  organization: NEC Labs America
– sequence: 2
  givenname: Ramin
  surname: Moslemi
  fullname: Moslemi, Ramin
  organization: NEC Labs America
– sequence: 3
  givenname: Francesco
  surname: Pittaluga
  fullname: Pittaluga, Francesco
  organization: NEC Labs America
– sequence: 4
  givenname: Buyu
  surname: Liu
  fullname: Liu, Buyu
  organization: NEC Labs America
– sequence: 5
  givenname: Manmohan
  surname: Chandraker
  fullname: Chandraker, Manmohan
  organization: NEC Labs America
BookMark eNotjN1KwzAYQKMouM09gV70BVK__HxJcyWj_kLBIdPbkbRfIaKpplXZ2zvRqwOHw5mzozQkYuxcQCkEuIv6ef2ojVa2lCBFCQJRH7C5MAa1RnDykM0kWuQWLJ6w5TjGACgBrHLVjF1exa_YEfep4_WQPj4pF_2Qi8Yn4qtvn6nYF5RHKjbZv1A7DXlXrDN1sZ3ikE7Zce9fR1r-c8Gebq439R1vHm7v61XDozRu4sHYvkcvAQGtC0FWVZAyaGi9MhbBB-v6XxnafdN3rkOp0Col0QuPXi3Y2d83EtH2Pcc3n3dbh9aiUeoH3LxK9Q
CODEN IEEPAD
ContentType Conference Proceeding
DBID 6IE
6IH
CBEJK
RIE
RIO
DOI 10.1109/CVPR46437.2021.01554
DatabaseName IEEE Electronic Library (IEL) Conference Proceedings
IEEE Proceedings Order Plan (POP) 1998-present by volume
IEEE Xplore All Conference Proceedings
IEEE Electronic Library Online
IEEE Proceedings Order Plans (POP) 1998-present
DatabaseTitleList
Database_xml – sequence: 1
  dbid: RIE
  name: IEEE Xplore
  url: https://proxy.k.utb.cz/login?url=https://ieeexplore.ieee.org/
  sourceTypes: Publisher
DeliveryMethod fulltext_linktorsrc
EISBN 1665445092
9781665445092
EISSN 2575-7075
EndPage 15803
ExternalDocumentID 9577563
Genre orig-research
GroupedDBID 6IE
6IH
6IL
6IN
ABLEC
ADZIZ
ALMA_UNASSIGNED_HOLDINGS
BEFXN
BFFAM
BGNUA
BKEBE
BPEOZ
CBEJK
CHZPO
IEGSK
IJVOP
OCL
RIE
RIL
RIO
ID FETCH-LOGICAL-i269t-b67ff5a2050579bb288b22b40ca36750ab79f288bbc205fd9d523573325a1a5a3
IEDL.DBID RIE
IngestDate Wed Jun 26 19:25:40 EDT 2024
IsPeerReviewed false
IsScholarly true
Language English
LinkModel DirectLink
MergedId FETCHMERGED-LOGICAL-i269t-b67ff5a2050579bb288b22b40ca36750ab79f288bbc205fd9d523573325a1a5a3
PageCount 10
ParticipantIDs ieee_primary_9577563
PublicationCentury 2000
PublicationDate 2021-01-01
PublicationDateYYYYMMDD 2021-01-01
PublicationDate_xml – month: 01
  year: 2021
  text: 2021-01-01
  day: 01
PublicationDecade 2020
PublicationTitle 2021 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR)
PublicationTitleAbbrev CVPR
PublicationYear 2021
Publisher IEEE
Publisher_xml – name: IEEE
SSID ssib052007398
ssib042469789
Score 2.4575782
Snippet Trajectory prediction is a safety-critical tool for autonomous vehicles to plan and execute actions. Our work addresses two key challenges in trajectory...
SourceID ieee
SourceType Publisher
StartPage 15794
SubjectTerms Benchmark testing
Computer vision
Context-aware services
Couplings
Semantics
Trajectory
Title Divide-and-Conquer for Lane-Aware Diverse Trajectory Prediction
URI https://ieeexplore.ieee.org/document/9577563
hasFullText 1
inHoldings 1
isFullTextHit
isPrint
link http://utb.summon.serialssolutions.com/2.0.0/link/0/eLvHCXMwjV3NS8MwFA_bTp5UNvGbHjyark2TpjmJbI4hToY42W0k7Suo0I3RIfrX-167zQ88eCshDflo-nufv8fYBSRaBwCOS-kkl3lmuQWRc0nxFyECgsnINDC6j4cTeTtV0wa73ObCAEAVfAY-PVa-_GyershU1jVKaxVHTdbUJq5ztTbfjhSo531jTic2IR2ZZJ0tFwam23saP0jyU6FWKEKfhAX5o6ZKBSmDXTbaTKaOJHn1V6Xz049fPI3_ne0e63wl73njLSztswYUbXbVp7Qr4LbIOPZDMFh6KK56d7YAfv1ml-D1qwgN8BC9XipT_jsOQ24cOroOmwxuHntDvq6dwJ9FbEruYp3nygoqVKeNcyJJnBBOBqmNUEcIrNMmp0aXYp88M5ki4psoEsqGVtnogLWKeQGHzEMFKZcoVkV4daXC94LQxnEKoQKN_wN3xNq0-NmipseYrdd9_HfzCduh7a-tGKesVS5XcIa4Xrrz6kA_AcsCn4s
link.rule.ids 309,310,780,784,789,790,796,23930,23931,25140,27925,54758
linkProvider IEEE
linkToHtml http://utb.summon.serialssolutions.com/2.0.0/link/0/eLvHCXMwjV3NS8MwFA9zHvSksonf9uDRdG2aNM1JZHNM3caQTXYbSfsKKnRSOkT_el_abX7gwVsJaUl4TX7v8_cIuYBISg_AUM4NpzxNNNXAUspt_oWPgKAS6xoYDMPehN9NxbRGLte1MABQJp-Bax_LWH4yjxfWVdZSQkoRBhtkU3DUc6tqrdXfwxlaet-40y2fkAxUtKyX8z3Vaj-OHriNVKFdyHzXqgv8R1eVElS6O2SwWk6VS_LiLgrjxh-_mBr_u95d0vwq33NGa2DaIzXIGuSqYwuvgOosoTgP4SB3UGF1-joDev2mc3A6ZY4GOIhfz6Uz_x0_YwM5VnhNMunejNs9uuyeQJ9YqApqQpmmQjPbqk4qY1gUGcYM92IdoJXgaSNVagdNjHPSRCXCUt8EARPa10IH-6SezTM4IA6aSClHxSrAw8sFvuf5Ogxj8AVIvBHMIWnYzc9eK4KM2XLfR38Pn5Ot3njQn_Vvh_fHZNuKovJpnJB6kS_gFFG-MGelcD8BZ0ui3g
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=proceeding&rft.title=2021+IEEE%2FCVF+Conference+on+Computer+Vision+and+Pattern+Recognition+%28CVPR%29&rft.atitle=Divide-and-Conquer+for+Lane-Aware+Diverse+Trajectory+Prediction&rft.au=Narayanan%2C+Sriram&rft.au=Moslemi%2C+Ramin&rft.au=Pittaluga%2C+Francesco&rft.au=Liu%2C+Buyu&rft.date=2021-01-01&rft.pub=IEEE&rft.eissn=2575-7075&rft.spage=15794&rft.epage=15803&rft_id=info:doi/10.1109%2FCVPR46437.2021.01554&rft.externalDocID=9577563