{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,3,15]],"date-time":"2026-03-15T07:35:17Z","timestamp":1773560117502,"version":"3.50.1"},"reference-count":103,"publisher":"Institute of Electrical and Electronics Engineers (IEEE)","issue":"1","license":[{"start":{"date-parts":[[2026,1,1]],"date-time":"2026-01-01T00:00:00Z","timestamp":1767225600000},"content-version":"vor","delay-in-days":0,"URL":"https:\/\/ieeexplore.ieee.org\/Xplorehelp\/downloads\/license-information\/IEEE.html"},{"start":{"date-parts":[[2026,1,1]],"date-time":"2026-01-01T00:00:00Z","timestamp":1767225600000},"content-version":"stm-asf","delay-in-days":0,"URL":"https:\/\/doi.org\/10.15223\/policy-029"},{"start":{"date-parts":[[2026,1,1]],"date-time":"2026-01-01T00:00:00Z","timestamp":1767225600000},"content-version":"stm-asf","delay-in-days":0,"URL":"https:\/\/doi.org\/10.15223\/policy-037"}],"content-domain":{"domain":[],"crossmark-restriction":false},"short-container-title":["IEEE Trans. Pattern Anal. Mach. Intell."],"published-print":{"date-parts":[[2026,1]]},"DOI":"10.1109\/tpami.2025.3611795","type":"journal-article","created":{"date-parts":[[2025,9,22]],"date-time":"2025-09-22T17:44:23Z","timestamp":1758563063000},"page":"914-931","source":"Crossref","is-referenced-by-count":1,"title":["Toward Optimal Mixture of Experts System for 3D Object Detection: A Game of Accuracy, Efficiency and Adaptivity"],"prefix":"10.1109","volume":"48","author":[{"ORCID":"https:\/\/orcid.org\/0009-0009-3749-2233","authenticated-orcid":false,"given":"Linshen","family":"Liu","sequence":"first","affiliation":[{"name":"Department of Civil and Systems Engineering (System Track), Johns Hopkins University, Baltimore, MD, USA"}],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Pu","family":"Wang","sequence":"additional","affiliation":[{"name":"Department of Civil and Systems Engineering (System Track), Johns Hopkins University, Baltimore, MD, USA"}],"role":[{"role":"author","vocabulary":"crossref"}]},{"ORCID":"https:\/\/orcid.org\/0009-0001-2799-9876","authenticated-orcid":false,"given":"Guanlin","family":"Wu","sequence":"additional","affiliation":[{"name":"Department of Civil and Systems Engineering (System Track), Johns Hopkins University, Baltimore, MD, USA"}],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Junyue","family":"Jiang","sequence":"additional","affiliation":[{"name":"Department of Civil and Systems Engineering (System Track), Johns Hopkins University, Baltimore, MD, USA"}],"role":[{"role":"author","vocabulary":"crossref"}]},{"ORCID":"https:\/\/orcid.org\/0000-0001-6431-8956","authenticated-orcid":false,"given":"Hao Frank","family":"Yang","sequence":"additional","affiliation":[{"name":"Department of Civil and Systems Engineering, Data Science and AI Institute, Johns Hopkins University, Baltimore, MD, USA"}],"role":[{"role":"author","vocabulary":"crossref"}]}],"member":"263","reference":[{"key":"ref1","doi-asserted-by":"publisher","DOI":"10.1109\/iccv.2019.00066"},{"key":"ref2","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR42600.2020.01164"},{"key":"ref3","first-page":"7482","article-title":"Multi-task learning using uncertainty to weigh losses for scene geometry and semantics","volume-title":"Proc. IEEE Conf. Comput. Vis. Pattern Recognit.","author":"Kendall"},{"key":"ref4","doi-asserted-by":"publisher","DOI":"10.1038\/s41591-018-0316-z"},{"key":"ref5","article-title":"Mobilenets: Efficient convolutional neural networks for mobile vision applications","author":"Howard","year":"2017"},{"key":"ref6","first-page":"6105","article-title":"Efficientnet: Rethinking model scaling for convolutional neural networks","volume-title":"Proc. Int. Conf. Mach. Learn.","author":"Tan"},{"key":"ref7","article-title":"ONNX: Open neural network exchange","author":"Bai","year":"2019"},{"key":"ref8","doi-asserted-by":"publisher","DOI":"10.1145\/3007787.3001163"},{"key":"ref9","first-page":"1387","article-title":"Dynamic network surgery for efficient DNNs","volume-title":"Proc. Adv. Neural Inf. Process. Syst.","author":"Guo"},{"key":"ref10","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR.2012.6248074"},{"key":"ref11","doi-asserted-by":"publisher","DOI":"10.1109\/TITS.2019.2892405"},{"key":"ref12","doi-asserted-by":"publisher","DOI":"10.1109\/TNNLS.2020.3015992"},{"key":"ref13","article-title":"GShard: Scaling giant models with conditional computation and automatic sharding","author":"Lepikhin","year":"2020"},{"key":"ref14","doi-asserted-by":"crossref","DOI":"10.36227\/techrxiv.172055626.64129172\/v1","article-title":"A survey on mixture of experts","author":"Cai","year":"2024"},{"key":"ref15","article-title":"Multi-modal sensor fusion for auto driving perception: A survey","author":"Huang","year":"2022"},{"key":"ref16","article-title":"GShard: Scaling giant models with conditional computation and automatic sharding","author":"Lepikhin","year":"2020"},{"key":"ref17","first-page":"5547","article-title":"Glam: Efficient scaling of language models with mixture-of-experts","volume-title":"Proc. Int. Conf. Mach. Learn.","author":"Du"},{"key":"ref18","first-page":"7103","article-title":"Mixture-of-experts with expertchoice routing","volume-title":"Proc. Adv. Neural Inf. Process. Syst.","author":"Zhou"},{"key":"ref19","first-page":"9564","article-title":"Multimodal contrastive learning with limoe: The language-image mixture of experts","volume-title":"Proc. Adv. Neural Inf. Process. Syst.","author":"Mustafa"},{"key":"ref20","first-page":"42048","article-title":"Mome: Mixture of multimodal experts for generalist multimodal large language models","volume-title":"Proc. Adv. Neural Inf. Process. Syst.","author":"Shen"},{"key":"ref21","article-title":"MoE-LLaVA: Mixture of experts for large vision-language models","author":"Lin","year":"2024"},{"key":"ref22","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR.2018.00472"},{"key":"ref23","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR.2019.00086"},{"key":"ref24","first-page":"9774","article-title":"Fast point R-CNN","volume-title":"Proc. 2019 IEEE\/CVF Int. Conf. Comput. Vis.","author":"Chen"},{"key":"ref25","doi-asserted-by":"publisher","DOI":"10.1109\/TPAMI.2020.2977026"},{"key":"ref26","doi-asserted-by":"publisher","DOI":"10.1609\/aaai.v35i2.16207"},{"key":"ref27","first-page":"18442","article-title":"Unifying voxel-based representation with transformer for 3D object detection","volume-title":"Adv. Neural Inf. Process. Syst.","author":"Li","year":"2022"},{"key":"ref28","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR52688.2022.00535"},{"key":"ref29","doi-asserted-by":"publisher","DOI":"10.1109\/ICRA48891.2023.10160968"},{"key":"ref30","article-title":"NVIDIA TensorRT developer guide","year":"2024"},{"key":"ref31","article-title":"XLA: Compiling machine learning for peak performance","author":"Sabne","year":"2020"},{"key":"ref32","first-page":"578","article-title":"$\\lbrace${TVM $\\rbrace$}: An automated $\\lbrace${ End-to-End$\\rbrace$} optimizing compiler for deep learning","volume-title":"Proc. 13th USENIX Symp. Operating Syst. Des. Implementation","author":"Chen","year":"2018"},{"key":"ref33","doi-asserted-by":"publisher","DOI":"10.1109\/TITS.2024.3439557"},{"key":"ref34","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR.2017.198"},{"key":"ref35","doi-asserted-by":"publisher","DOI":"10.1007\/978-3-030-58580-8_38"},{"key":"ref36","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR52729.2023.02050"},{"key":"ref37","doi-asserted-by":"publisher","DOI":"10.1109\/iccv51070.2023.00840"},{"key":"ref38","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR52729.2023.01264"},{"key":"ref39","doi-asserted-by":"publisher","DOI":"10.1007\/978-3-031-19769-7_5"},{"key":"ref40","doi-asserted-by":"publisher","DOI":"10.1109\/WACV57701.2024.00737"},{"key":"ref41","doi-asserted-by":"publisher","DOI":"10.1609\/aaai.v38i6.28436"},{"key":"ref42","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR46437.2021.00052"},{"key":"ref43","doi-asserted-by":"publisher","DOI":"10.1016\/j.trc.2022.103982"},{"key":"ref44","doi-asserted-by":"publisher","DOI":"10.1109\/TPAMI.2023.3300976"},{"key":"ref45","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR52729.2023.02099"},{"key":"ref46","doi-asserted-by":"publisher","DOI":"10.1109\/TPAMI.2023.3335480"},{"key":"ref47","first-page":"22158","article-title":"Hierarchical neural architecture search for deep stereo matching","volume-title":"Proc. Adv. Neural Inf. Process. Syst.","author":"Cheng"},{"key":"ref48","doi-asserted-by":"publisher","DOI":"10.1109\/TITS.2022.3154368"},{"key":"ref49","first-page":"21224","article-title":"Every view counts: Cross-view consistency in 3D object detection with hybrid-cylindrical-spherical voxelization","volume-title":"Proc. 34th Int. Conf. Neural Inf. Process. Syst.","author":"Chen"},{"key":"ref50","doi-asserted-by":"publisher","DOI":"10.1109\/tpami.2020.2977026"},{"key":"ref51","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR.2019.01298"},{"key":"ref52","doi-asserted-by":"publisher","DOI":"10.1109\/IROS.2017.8205955"},{"issue":"10","key":"ref53","doi-asserted-by":"crossref","DOI":"10.3390\/s18103337","article-title":"SECOND: Sparsely embedded convolutional detection","volume":"18","author":"Yan","year":"2018","journal-title":"Sensors"},{"key":"ref54","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR42600.2020.01054"},{"key":"ref55","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR42600.2020.01105"},{"key":"ref56","doi-asserted-by":"publisher","DOI":"10.1007\/978-3-030-58542-6_2"},{"key":"ref57","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR46437.2021.01161"},{"key":"ref58","doi-asserted-by":"publisher","DOI":"10.1007\/978-3-030-58555-6_3"},{"key":"ref59","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR42600.2020.00466"},{"key":"ref60","doi-asserted-by":"publisher","DOI":"10.1109\/WACV48630.2021.00157"},{"key":"ref61","doi-asserted-by":"publisher","DOI":"10.1109\/TCSS.2023.3286543"},{"key":"ref62","doi-asserted-by":"publisher","DOI":"10.1109\/IVS.2018.8500387"},{"key":"ref63","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR.2018.00033"},{"key":"ref64","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR52729.2023.01296"},{"key":"ref65","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR52688.2022.01667"},{"key":"ref66","doi-asserted-by":"publisher","DOI":"10.1109\/TNNLS.2023.3325527"},{"key":"ref67","doi-asserted-by":"publisher","DOI":"10.1109\/tpami.2025.3565194"},{"key":"ref68","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR52688.2022.00534"},{"key":"ref69","doi-asserted-by":"publisher","DOI":"10.1109\/ICCV51070.2023.01675"},{"key":"ref70","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR52729.2023.02074"},{"key":"ref71","doi-asserted-by":"publisher","DOI":"10.1109\/TITS.2023.3347078"},{"key":"ref72","article-title":"RoboFusion: Towards robust multi-modal 3D obiect detection via SAM","author":"Song","year":"2024"},{"key":"ref73","doi-asserted-by":"publisher","DOI":"10.1109\/TCSVT.2023.3306361"},{"key":"ref74","article-title":"Onnx: Open neural network exchange","author":"Community","year":"2024"},{"key":"ref75","article-title":"MLIR: A compiler infrastructure for the end of moore\u2019s law","author":"Lattner","year":"2020"},{"key":"ref76","doi-asserted-by":"publisher","DOI":"10.1109\/CGO.2004.1281665"},{"key":"ref77","doi-asserted-by":"publisher","DOI":"10.1109\/RusAutoCon52004.2021.9537452"},{"key":"ref78","doi-asserted-by":"publisher","DOI":"10.1109\/CGO57630.2024.10444871"},{"key":"ref79","first-page":"10215","article-title":"Glow: Generative flow with invertible 1x1 convolutions","volume-title":"Proc. Adv. Neural Inf. Process. Syst.","author":"Kingma"},{"key":"ref80","doi-asserted-by":"publisher","DOI":"10.1145\/3620665.3640366"},{"key":"ref81","article-title":"BEVDet: High-performance multi-camera 3D object detection in bird-eye-view","author":"Huang","year":"2021"},{"key":"ref82","doi-asserted-by":"publisher","DOI":"10.1162\/neco.1991.3.1.79"},{"key":"ref83","doi-asserted-by":"publisher","DOI":"10.1109\/ICCV48922.2021.00018"},{"key":"ref84","doi-asserted-by":"publisher","DOI":"10.5244\/C.29.150"},{"key":"ref85","doi-asserted-by":"publisher","DOI":"10.1145\/248208.237198"},{"key":"ref86","volume-title":"Communication Complexity and Parallel Computing","author":"Hromkovi\u010d","year":"2013"},{"key":"ref87","first-page":"145","article-title":"Arachne:$\\lbrace${Core-Aware$\\rbrace$} thread management","volume-title":"Proc. 13th USENIX Symp. Operating Syst. Des. Implementation","author":"Qin"},{"key":"ref88","doi-asserted-by":"publisher","DOI":"10.1109\/ISCA52012.2021.00090"},{"key":"ref89","doi-asserted-by":"publisher","DOI":"10.1109\/WACV51458.2022.00235"},{"key":"ref90","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR42600.2020.01054"},{"key":"ref91","doi-asserted-by":"publisher","DOI":"10.1007\/s44196-021-00056-3"},{"key":"ref92","first-page":"2743","article-title":"Improving 3D object detection with channel-wise transformer","volume-title":"Proc. IEEE\/CVF Int. Conf. Comput. Vis.","author":"Sheng"},{"key":"ref93","doi-asserted-by":"publisher","DOI":"10.1609\/aaai.v37i3.25380"},{"key":"ref94","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR52729.2023.01681"},{"key":"ref95","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR52688.2022.00119"},{"key":"ref96","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR52729.2023.02076"},{"key":"ref97","article-title":"Sparse4D v3: Advancing end-to-end 3D detection and tracking","author":"Lin","year":"2023"},{"key":"ref98","article-title":"RCBEVDet: Toward high-accuracy radar-camera fusion 3D perception network","author":"Lin","year":"2024"},{"key":"ref99","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR52688.2022.00116"},{"key":"ref100","doi-asserted-by":"publisher","DOI":"10.1609\/aaai.v39i10.33140"},{"key":"ref101","article-title":"ImOV3D: Learning open-vocabulary point clouds 3D object detection from only 2D images","volume-title":"Proc. Int. Conf. Neural Inf. Process. Syst.","author":"Yang"},{"key":"ref102","doi-asserted-by":"publisher","DOI":"10.1109\/JIOT.2018.2876298"},{"key":"ref103","doi-asserted-by":"publisher","DOI":"10.1109\/TVT.2019.2959410"}],"container-title":["IEEE Transactions on Pattern Analysis and Machine Intelligence"],"original-title":[],"link":[{"URL":"http:\/\/xplorestaging.ieee.org\/ielx8\/34\/11275622\/11175038.pdf?arnumber=11175038","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2025,12,4]],"date-time":"2025-12-04T21:01:36Z","timestamp":1764882096000},"score":1,"resource":{"primary":{"URL":"https:\/\/ieeexplore.ieee.org\/document\/11175038\/"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2026,1]]},"references-count":103,"journal-issue":{"issue":"1"},"URL":"https:\/\/doi.org\/10.1109\/tpami.2025.3611795","relation":{},"ISSN":["0162-8828","2160-9292","1939-3539"],"issn-type":[{"value":"0162-8828","type":"print"},{"value":"2160-9292","type":"electronic"},{"value":"1939-3539","type":"electronic"}],"subject":[],"published":{"date-parts":[[2026,1]]}}}