{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2025,4,11]],"date-time":"2025-04-11T10:10:01Z","timestamp":1744366201499,"version":"3.40.4"},"reference-count":35,"publisher":"IOP Publishing","issue":"2","license":[{"start":{"date-parts":[[2025,4,11]],"date-time":"2025-04-11T00:00:00Z","timestamp":1744329600000},"content-version":"vor","delay-in-days":0,"URL":"https:\/\/creativecommons.org\/licenses\/by\/4.0\/"},{"start":{"date-parts":[[2025,4,11]],"date-time":"2025-04-11T00:00:00Z","timestamp":1744329600000},"content-version":"tdm","delay-in-days":0,"URL":"https:\/\/iopscience.iop.org\/info\/page\/text-and-data-mining"}],"funder":[{"DOI":"10.13039\/501100013061","name":"Jilin Provincial Scientific and Technological Development Program","doi-asserted-by":"crossref","award":["20230101174JC"],"award-info":[{"award-number":["20230101174JC"]}],"id":[{"id":"10.13039\/501100013061","id-type":"DOI","asserted-by":"crossref"}]}],"content-domain":{"domain":["iopscience.iop.org"],"crossmark-restriction":false},"short-container-title":["Mach. Learn.: Sci. Technol."],"published-print":{"date-parts":[[2025,6,30]]},"abstract":"<jats:title>Abstract<\/jats:title>\n               <jats:p>To solve the problem that the lack of a global perspective leads to local misestimation and overall structural dislocation when optical flow estimates large-scale motion and complex scenes, this paper proposes an optical flow estimation based on global cross information and dynamic encoder\u2013dynamic decoder. The network architecture is improved stage by stage according to the streaming direction of the encoder and decoder data streams. For the encoder, relative and absolute position coding are adopted to construct a mixed coding network, and the learnable weights are used to dynamically adjust the mixed position coding information to enrich the global and local position information. For the enhancer layer after the encoder, the contextual information of the input sequence is captured through cross attention and a feed-forward neural network to construct a global cross information attention network, which acquires the global perspective under the premise of adapting to large-scale motion. For the decoder, bilinear interpolation and deformable convolution are combined to construct a dynamic anisotropic upsampling module, and dynamic anisotropic upsampling of the input feature maps is realized by adjusting the offsets of the sampling points to enhance the ability to process high-resolution details and complex motion boundaries. Finally, the performance of the proposed method is evaluated in the endpoint error value, estimation time and number of parameters. The experimental results show that compared with the RAFT benchmark network, the model in this paper preserves the global features and edge detail information of large-scale motion without significantly increasing the number of model parameters.<\/jats:p>","DOI":"10.1088\/2632-2153\/adc8fa","type":"journal-article","created":{"date-parts":[[2025,4,3]],"date-time":"2025-04-03T22:54:57Z","timestamp":1743720897000},"page":"025011","update-policy":"https:\/\/doi.org\/10.1088\/crossmark-policy","source":"Crossref","is-referenced-by-count":0,"title":["Optical flow estimation based on global cross information and dynamic encoder\u2013dynamic decoder"],"prefix":"10.1088","volume":"6","author":[{"given":"Haoxin","family":"Guo","sequence":"first","affiliation":[],"role":[{"role":"author","vocabulary":"crossref"}]},{"ORCID":"https:\/\/orcid.org\/0000-0001-5056-2093","authenticated-orcid":true,"given":"Yifan","family":"Wang","sequence":"additional","affiliation":[],"role":[{"role":"author","vocabulary":"crossref"}]},{"ORCID":"https:\/\/orcid.org\/0009-0003-6398-8423","authenticated-orcid":true,"given":"Xiaobo","family":"Guo","sequence":"additional","affiliation":[],"role":[{"role":"author","vocabulary":"crossref"}]}],"member":"266","published-online":{"date-parts":[[2025,4,11]]},"reference":[{"key":"mlstadc8fabib1","doi-asserted-by":"publisher","first-page":"1731","DOI":"10.3390\/s23031731","article-title":"Vehicle detection and recognition approach in multi-scale traffic monitoring system via graph-based data optimization","volume":"23","author":"Wieczorek","year":"2023","journal-title":"Sensors"},{"key":"mlstadc8fabib2","doi-asserted-by":"publisher","first-page":"7630","DOI":"10.1038\/s41598-023-34777-6","article-title":"Dynamic obstacle detection method based on U\u2013V disparity and residual optical flow for autonomous driving","volume":"13","author":"Yuan","year":"2023","journal-title":"Sci. Rep."},{"key":"mlstadc8fabib3","doi-asserted-by":"publisher","first-page":"1819","DOI":"10.1007\/s11277-023-10682-z","article-title":"A comprehensive review of various approach for medical image segmentation and disease prediction","volume":"132","author":"Narayan","year":"2023","journal-title":"Wirel. Pers. Commun."},{"key":"mlstadc8fabib4","doi-asserted-by":"publisher","first-page":"2758","DOI":"10.48550\/arXiv.1504.06852","article-title":"FlowNet: learning optical flow with convolutional networks","author":"Dosovitskiy","year":"2015"},{"key":"mlstadc8fabib5","doi-asserted-by":"publisher","first-page":"2720","DOI":"10.1109\/cvpr.2017.291","article-title":"Optical flow estimation using a spatial pyramid network","author":"Anurag","year":"2017"},{"key":"mlstadc8fabib6","doi-asserted-by":"publisher","first-page":"234","DOI":"10.1007\/978-3-319-24574-4_28","article-title":"U-Net: convolutional networks for biomedical image segmentation","author":"Ronneberger","year":"2015"},{"key":"mlstadc8fabib7","doi-asserted-by":"publisher","first-page":"402","DOI":"10.1007\/978-3-030-58536-5_24","article-title":"RAFT: recurrent all-pairs field transforms for optical flow","volume":"vol 16","author":"Teed","year":"2020"},{"key":"mlstadc8fabib8","doi-asserted-by":"publisher","first-page":"1575","DOI":"10.14778\/3685800.3685803","article-title":"BRAFT: Baidu\u2019s Raft consistency algorithm implementation based on BRPC","volume":"28","author":"Jia","year":"2024","journal-title":"IEEE Signal Process. Lett."},{"key":"mlstadc8fabib9","doi-asserted-by":"publisher","first-page":"1835","DOI":"10.1007\/s11263-023-01930-7","article-title":"MS-RAFT+: high resolution multi-scale RAFT","volume":"132","author":"Jahedi","year":"2024","journal-title":"Int. J. Comput. Vis."},{"key":"mlstadc8fabib10","doi-asserted-by":"publisher","DOI":"10.1088\/1361-6501\/ad3183","article-title":"A method of applying deep learning based optical flow algorithm to river flow discharge measurement","volume":"35","author":"Wang","year":"2024","journal-title":"Meas. Sci. Technol."},{"key":"mlstadc8fabib11","doi-asserted-by":"publisher","DOI":"10.3390\/app132413056","article-title":"Real-time optical flow estimation method based on cross-stage network","volume":"13","author":"Park","year":"2023","journal-title":"Appl. Sci."},{"key":"mlstadc8fabib12","doi-asserted-by":"publisher","first-page":"1","DOI":"10.1007\/s00348-023-03695-8","article-title":"LightPIVNet: an effective convolutional neural network for particle image velocimetry","volume":"70","author":"Yu","year":"2021","journal-title":"IEEE Trans. Instrum. Meas."},{"key":"mlstadc8fabib13","doi-asserted-by":"publisher","first-page":"4639","DOI":"10.1109\/TCSVT.2021.3135440","article-title":"MRDFlow: unsupervised optical flow estimation network with multi-scale recurrent decoder","volume":"32","author":"Zhao","year":"2021","journal-title":"IEEE Trans. Circuits Syst. Video Technol."},{"key":"mlstadc8fabib14","doi-asserted-by":"publisher","first-page":"39007","DOI":"10.1364\/OE.505993","article-title":"Kinematic target surface sensing based on improved deep optical flow tracking","volume":"31","author":"Lu","year":"2023","journal-title":"Opt. Express"},{"key":"mlstadc8fabib15","doi-asserted-by":"publisher","first-page":"2462","DOI":"10.48550\/arXiv.1612.01925","article-title":"FlowNet 2.0: evolution of optical flow estimation with deep networks","author":"Ilg","year":"2017"},{"key":"mlstadc8fabib16","doi-asserted-by":"publisher","first-page":"614","DOI":"10.48550\/arXiv.1808.01838","article-title":"Occlusions, motion and depth boundaries with a generic network for disparity, optical flow or scene flow estimation","author":"Ilg","year":"2018"},{"key":"mlstadc8fabib17","doi-asserted-by":"publisher","first-page":"2566","DOI":"10.1109\/lra.2023.3254431","article-title":"PT-FlowNet: scene flow estimation on point clouds with point transformer","volume":"8","author":"Fu","year":"2023","journal-title":"IEEE Robot. Autom. Lett."},{"key":"mlstadc8fabib18","doi-asserted-by":"publisher","first-page":"5026","DOI":"10.48550\/arXiv.2106.04195","article-title":"Learning by distillation: a self-supervised learning framework for optical flow estimation","volume":"44","author":"Liu","year":"2022","journal-title":"IEEE Trans. Pattern Anal. Mach. Intell."},{"key":"mlstadc8fabib19","doi-asserted-by":"publisher","first-page":"398","DOI":"10.1007\/s10489-023-05131-0","article-title":"An optical flow estimation method based on multiscale anisotropic convolution","volume":"54","author":"Wang","year":"2024","journal-title":"Appl. Intell."},{"article-title":"Dynamic U-Net: adaptively calibrate features for abdominal multi-organ segmentation","year":"2024","author":"Yang","key":"mlstadc8fabib20"},{"key":"mlstadc8fabib21","doi-asserted-by":"publisher","DOI":"10.1016\/j.eswa.2023.123111","article-title":"MICU: image super-resolution via multi-level information compensation and U-net","volume":"245","author":"Chen","year":"2023","journal-title":"Expert Syst. Appl."},{"key":"mlstadc8fabib22","doi-asserted-by":"publisher","first-page":"677","DOI":"10.1109\/tcsvt.2022.3205375","article-title":"MDFlow: unsupervised optical flow learning by reliable mutual knowledge distillation","volume":"33","author":"Kong","year":"2023","journal-title":"IEEE Trans. Circuits Syst. Video Technol."},{"key":"mlstadc8fabib23","doi-asserted-by":"publisher","DOI":"10.1016\/j.autcon.2023.105138","article-title":"Asymmetric dual-decoder-U-Net for pavement crack semantic segmentation","volume":"156","author":"Zaid","year":"2023","journal-title":"Autom. Constr."},{"key":"mlstadc8fabib24","doi-asserted-by":"publisher","first-page":"21544","DOI":"10.1109\/JSEN.2023.3301187","article-title":"Seq2Dense U-Net: analyzing sequential inertial sensor data for human activity recognition using dense segmentation model","volume":"23","author":"Meena","year":"2023","journal-title":"IEEE Sens. J."},{"key":"mlstadc8fabib25","doi-asserted-by":"publisher","first-page":"5747","DOI":"10.48550\/arXiv.1904.05290","article-title":"Iterative residual refinement for joint optical flow and occlusion estimation","author":"Hur","year":"2019"},{"key":"mlstadc8fabib26","doi-asserted-by":"publisher","first-page":"1173","DOI":"10.1007\/s13042-020-01227-y","article-title":"Towards wide-scale continuous gesture recognition model for in-depth and grayscale input videos","volume":"12","author":"Mahmoud","year":"2021","journal-title":"Int. J. Mach. Learn. Cybern."},{"article-title":"Occinpflow: occlusion-inpainting optical flow estimation by unsupervised learning","year":"2020","author":"Luo","key":"mlstadc8fabib27"},{"key":"mlstadc8fabib28","doi-asserted-by":"publisher","first-page":"346","DOI":"10.1016\/j.neucom.2021.11.037","article-title":"FPCR-Net: feature pyramidal correlation and residual reconstruction for optical flow estimation","volume":"417","author":"Song","year":"2022","journal-title":"Neurocomputing"},{"key":"mlstadc8fabib29","doi-asserted-by":"publisher","first-page":"10433","DOI":"10.1109\/tits.2023.3276328","article-title":"PSNet: parallel pyramid pooling stereo network","volume":"24","author":"Emlek","year":"2023","journal-title":"IEEE Trans. Intell. Transp. Syst."},{"key":"mlstadc8fabib30","doi-asserted-by":"publisher","DOI":"10.1016\/j.knosys.2023.110568","article-title":"UDF-GAN: unsupervised dense optical-flow estimation using cycle generative adversarial networks","volume":"271","author":"Liu","year":"2023","journal-title":"Knowl.-Based Syst."},{"key":"mlstadc8fabib31","doi-asserted-by":"publisher","first-page":"1","DOI":"10.1109\/mvip62238.2024.10491183","article-title":"Rethinking RAFT for efficient optical flow","author":"Eslami","year":"2024"},{"key":"mlstadc8fabib32","doi-asserted-by":"publisher","first-page":"38275","DOI":"10.1109\/access.2023.3267635","article-title":"River surface velocimetry based on virtual river dataset and modulated-RAFT","volume":"11","author":"Wu","year":"2023","journal-title":"IEEE Access"},{"key":"mlstadc8fabib33","doi-asserted-by":"publisher","first-page":"1","DOI":"10.1109\/icrai57502.2023.10089609","article-title":"Video stabilization using RAFT-based optical flow","author":"Ashar","year":"2023"},{"key":"mlstadc8fabib34","doi-asserted-by":"publisher","first-page":"1817","DOI":"10.1002\/rob.22219","article-title":"Neural network\u2010based optical flow versus traditional optical flow techniques with thermal aerial imaging in real\u2010world settings","volume":"40","author":"Nguyen","year":"2023","journal-title":"J. Field Robot."},{"key":"mlstadc8fabib35","doi-asserted-by":"publisher","first-page":"6044","DOI":"10.48550\/arXiv.1812.06264","article-title":"Hierarchical discrete distribution decomposition for match density estimation","author":"Yin","year":"2019"}],"container-title":["Machine Learning: Science and Technology"],"original-title":[],"link":[{"URL":"https:\/\/iopscience.iop.org\/article\/10.1088\/2632-2153\/adc8fa","content-type":"text\/html","content-version":"am","intended-application":"text-mining"},{"URL":"https:\/\/iopscience.iop.org\/article\/10.1088\/2632-2153\/adc8fa\/pdf","content-type":"application\/pdf","content-version":"am","intended-application":"text-mining"},{"URL":"https:\/\/iopscience.iop.org\/article\/10.1088\/2632-2153\/adc8fa","content-type":"text\/html","content-version":"vor","intended-application":"text-mining"},{"URL":"https:\/\/iopscience.iop.org\/article\/10.1088\/2632-2153\/adc8fa\/pdf","content-type":"application\/pdf","content-version":"vor","intended-application":"text-mining"},{"URL":"https:\/\/iopscience.iop.org\/article\/10.1088\/2632-2153\/adc8fa\/pdf","content-type":"application\/pdf","content-version":"am","intended-application":"syndication"},{"URL":"https:\/\/iopscience.iop.org\/article\/10.1088\/2632-2153\/adc8fa\/pdf","content-type":"application\/pdf","content-version":"vor","intended-application":"syndication"},{"URL":"https:\/\/iopscience.iop.org\/article\/10.1088\/2632-2153\/adc8fa\/pdf","content-type":"application\/pdf","content-version":"am","intended-application":"similarity-checking"},{"URL":"https:\/\/iopscience.iop.org\/article\/10.1088\/2632-2153\/adc8fa\/pdf","content-type":"application\/pdf","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2025,4,11]],"date-time":"2025-04-11T09:28:32Z","timestamp":1744363712000},"score":1,"resource":{"primary":{"URL":"https:\/\/iopscience.iop.org\/article\/10.1088\/2632-2153\/adc8fa"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2025,4,11]]},"references-count":35,"journal-issue":{"issue":"2","published-online":{"date-parts":[[2025,4,11]]},"published-print":{"date-parts":[[2025,6,30]]}},"URL":"https:\/\/doi.org\/10.1088\/2632-2153\/adc8fa","relation":{},"ISSN":["2632-2153"],"issn-type":[{"type":"electronic","value":"2632-2153"}],"subject":[],"published":{"date-parts":[[2025,4,11]]},"assertion":[{"value":"Optical flow estimation based on global cross information and dynamic encoder\u2013dynamic decoder","name":"article_title","label":"Article Title"},{"value":"Machine Learning: Science and Technology","name":"journal_title","label":"Journal Title"},{"value":"paper","name":"article_type","label":"Article Type"},{"value":"\u00a9 2025 The Author(s). Published by IOP Publishing Ltd","name":"copyright_information","label":"Copyright Information"},{"value":"2024-08-13","name":"date_received","label":"Date Received","group":{"name":"publication_dates","label":"Publication dates"}},{"value":"2025-04-03","name":"date_accepted","label":"Date Accepted","group":{"name":"publication_dates","label":"Publication dates"}},{"value":"2025-04-11","name":"date_epub","label":"Online publication date","group":{"name":"publication_dates","label":"Publication dates"}}]}}