001035216 001__ 1035216
001035216 005__ 20250203124500.0
001035216 0247_ $$2doi$$a10.1109/CVPR52733.2024.00149
001035216 0247_ $$2datacite_doi$$a10.34734/FZJ-2025-00301
001035216 0247_ $$2WOS$$aWOS:001322555901082
001035216 037__ $$aFZJ-2025-00301
001035216 041__ $$aEnglish
001035216 1001_ $$0P:(DE-HGF)0$$aYue, Jiangbei$$b0
001035216 1112_ $$a2024 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR)$$cSeattle$$d2024-06-16 - 2024-06-22$$wWA
001035216 245__ $$aHuman Motion Prediction Under Unexpected Perturbation
001035216 260__ $$bIEEE$$c2024
001035216 300__ $$a1501-1511
001035216 3367_ $$2ORCID$$aCONFERENCE_PAPER
001035216 3367_ $$033$$2EndNote$$aConference Paper
001035216 3367_ $$2BibTeX$$aINPROCEEDINGS
001035216 3367_ $$2DRIVER$$aconferenceObject
001035216 3367_ $$2DataCite$$aOutput Types/Conference Paper
001035216 3367_ $$0PUB:(DE-HGF)8$$2PUB:(DE-HGF)$$aContribution to a conference proceedings$$bcontrib$$mcontrib$$s1737998262_9342
001035216 520__ $$aWe investigate a new task in human motion prediction, which is predicting motions under unexpected physical perturbation potentially involving multiple people. Compared with existing research, this task involves predicting less controlled, unpremeditated and pure reactive motions in response to external impact and how such motions can propagate through people. It brings new challenges such as data scarcity and predicting complex interactions. To this end, we propose a new method capitalizing differentiable physics and deep neural networks, leading to an explicit Latent Differentiable Physics (LDP) model. Through experiments, we demonstrate that LDP has high data efficiency, outstanding prediction accuracy, strong generalizability and good explainability. Since there is no similar research, a comprehensive comparison with 11 adapted baselines from several relevant domains is conducted, showing LDP outperforming existing research both quantitatively and qualitatively, improving prediction accuracy by as much as 70%, and demonstrating significantly stronger generalization.
001035216 536__ $$0G:(DE-HGF)POF4-5111$$a5111 - Domain-Specific Simulation & Data Life Cycle Labs (SDLs) and Research Groups (POF4-511)$$cPOF4-511$$fPOF IV$$x0
001035216 536__ $$0G:(EU-Grant)899739$$aCrowdDNA - TECHNOLOGIES FOR COMPUTER-ASSISTED CROWD MANAGEMENT (899739)$$c899739$$fH2020-FETOPEN-2018-2019-2020-01$$x1
001035216 588__ $$aDataset connected to CrossRef Conference
001035216 7001_ $$0P:(DE-HGF)0$$aLi, Baiyi$$b1
001035216 7001_ $$0P:(DE-HGF)0$$aPetrré, Julien$$b2
001035216 7001_ $$0P:(DE-Juel1)132266$$aSeyfried, Armin$$b3$$ufzj
001035216 7001_ $$0P:(DE-HGF)0$$aWang, He$$b4$$eCorresponding author
001035216 773__ $$a10.1109/CVPR52733.2024.00149
001035216 8564_ $$uhttps://juser.fz-juelich.de/record/1035216/files/2403.15891v1.pdf$$yOpenAccess
001035216 909CO $$ooai:juser.fz-juelich.de:1035216$$pdnbdelivery$$pec_fundedresources$$pVDB$$pdriver$$popen_access$$popenaire
001035216 9101_ $$0I:(DE-588b)5008462-8$$6P:(DE-Juel1)132266$$aForschungszentrum Jülich$$b3$$kFZJ
001035216 9101_ $$0I:(DE-HGF)0$$6P:(DE-HGF)0$$aExternal Institute$$b4$$kExtern
001035216 9131_ $$0G:(DE-HGF)POF4-511$$1G:(DE-HGF)POF4-510$$2G:(DE-HGF)POF4-500$$3G:(DE-HGF)POF4$$4G:(DE-HGF)POF$$9G:(DE-HGF)POF4-5111$$aDE-HGF$$bKey Technologies$$lEngineering Digital Futures – Supercomputing, Data Management and Information Security for Knowledge and Action$$vEnabling Computational- & Data-Intensive Science and Engineering$$x0
001035216 9141_ $$y2024
001035216 915__ $$0StatID:(DE-HGF)0510$$2StatID$$aOpenAccess
001035216 920__ $$lyes
001035216 9201_ $$0I:(DE-Juel1)IAS-7-20180321$$kIAS-7$$lZivile Sicherheitsforschung$$x0
001035216 980__ $$acontrib
001035216 980__ $$aVDB
001035216 980__ $$aUNRESTRICTED
001035216 980__ $$aI:(DE-Juel1)IAS-7-20180321
001035216 9801_ $$aFullTexts