 二维码(扫一下试试看!) | 基于特征匹配的非刚性大位移光流算法 | Non-Rigid and Large Displacement Optical Flow Based on Descriptor Matching | 投稿时间:2019-01-15 | DOI:10.15918/j.tbit1001-0645.2019.023 | 中文关键词:特征匹配光流大位移MPI Sintel | English Keywords:descriptor matchingoptical flowlarge displacementMPI Sintel | 基金项目:国家部委预研基金资助项目(51327030104) | | 摘要点击次数:805 | 全文下载次数:315 | 中文摘要: | 针对非刚性大位移光流估计问题,提出了一种基于特征匹配的光流优化算法. 首先,传统数据项针对整副图像进行一致性假设的策略过于粗放,文中采用了一种根据图像中每个像素特点自适应选择一致性假设的优化方法;其次,针对传统各向同性的平滑项易造成演化过程中图像细节丢失、边缘模糊等问题,提出一种各向异性的平滑项,有利于实现图像细节的光流估计;另外融合特征匹配于变分光流框架中,充分利用特征匹配在大位移条件下的鲁棒性与变分光流的致密性;最后,基于MPI Sintel数据集对本文算法进行定量分析. 实验结果表明,本文算法能够实现非刚性大位移光流的准确估计,鲁棒性强,在market_5图像中,该算法比LDOF于AAE和AEPE指标上分别提升了18.9%与21.9%. | English Summary: | In order to solve the arithmetic problem of non-rigid and large displacement optical flow, an improved arithmetic was proposed based on descriptor matching for optical flow. In the method, a novel data term was put forward firstly to adaptively select invariable hypothesizes for each pixel. And then, the traditional isotropic smoothness term was optimized to be the anisotropic one for the optical flow estimation of image details. Furthermore, the descriptor matching was applied to optical flow fields, taking advantage of the robust ability of descriptor matching to produce some large displacement correspondences and the ability of dense optical flow. At last, a quantitative analysis of the approach was performed on MPI Sintel.The results show that the proposed method can realize accurate estimation of non-rigid and large displacement optical flow. And the method is superior to LDOF with a relative gain of 18.9% for AAE and 21.9% for AEPE in market_5. | 查看全文查看/发表评论下载PDF阅读器 | |
邢伯阳,潘峰,冯肖雪.ArUco-SLAM:基于ArUco二维码阵列的单目实时建图定位系统[J].北京理工大学学报(自然科学版),2020,40(4):427~433.XINGBo-yang,PANFeng,FENGXiao-xue.ArUco-SLAM:aMonocularSLAMSystemBa ... 北京理工大学科研学术 本站小编 Free考研考试 2021-12-21付波,谢安桓,赵春宇,周华.电液伺服随机振动控制系统非线性建模与仿真[J].北京理工大学学报(自然科学版),2020,40(5):491~495.FUBo,XIEAn-huan,ZHAOChun-yu,ZHOUHua.NonlinearModelingandSimulationonElectro-H ... 北京理工大学科研学术 本站小编 Free考研考试 2021-12-21AUNGKhaingZaw,刘锴,蒋立宁.双体量子系统的算子权并发度[J].北京理工大学学报(自然科学版),2020,40(6):687~690.AUNGKhaingZaw,LIUKai,JIANGLi-ning.OperatorWeightedConcurrenceMeasureonTwo-Qub ... 北京理工大学科研学术 本站小编 Free考研考试 2021-12-21武浩,张付军,王政凯,高宏力.航空二冲程直喷发动机电控润滑系统试验研究[J].北京理工大学学报(自然科学版),2020,40(7):719~724.WUHao,ZHANGFu-jun,WANGZheng-kai,GAOHong-li.ExperimentalResearchonElectronicC ... 北京理工大学科研学术 本站小编 Free考研考试 2021-12-21毛婷,梁玮.自动驾驶算法的异常事件生成系统设计与评估[J].北京理工大学学报(自然科学版),2020,40(7):753~759.MAOTing,LIANGWei.TheDesignandEvaluationofAbnormalEventGenerationSystemforAutonomousDr ... 北京理工大学科研学术 本站小编 Free考研考试 2021-12-21严雪飞,刘卫东,王佳,欧渊,石根柱.GTEM室频域校准系统及其不确定度评定[J].北京理工大学学报(自然科学版),2020,40(7):771~777.YANXue-fei,LIUWei-dong,WANGJia,OUYuan,SHIGen-zhu.FrequencyDomainCalibratio ... 北京理工大学科研学术 本站小编 Free考研考试 2021-12-21马立玲,郭凯杰,王军政.基于改进SVM的车辆传动系统故障诊断方法[J].北京理工大学学报(自然科学版),2020,40(8):856~860.MALi-ling,GUOKai-jie,WANGJun-zheng.AFaultDiagnosisMethodofVehicleTransmissionSy ... 北京理工大学科研学术 本站小编 Free考研考试 2021-12-21李明,刘若愚,姜春兰,寇家勋.基于无线智能节点的远程控制系统设计与实现[J].北京理工大学学报(自然科学版),2020,40(8):849~855.LIMing,LIURuo-yu,JIANGChun-lan,KOUJia-xun.DesignoftheRemote-ControlSystemfor ... 北京理工大学科研学术 本站小编 Free考研考试 2021-12-21吴志东,房俊龙,何义波,潘迪,李强.压电能量俘获系统建模及仿真[J].北京理工大学学报(自然科学版),2020,40(8):894~900.WUZhi-dong,FANGJun-long,HEYi-bo,PANDi,LIQiang.ModelingandSimulationofPiezoelectr ... 北京理工大学科研学术 本站小编 Free考研考试 2021-12-21李贵强,吴正斌,姜帆,林祥辉,武利强.基于驱动电机控制的电动汽车坡道静止保持系统[J].北京理工大学学报(自然科学版),2020,40(9):941~948.LIGui-qiang,WUZheng-bin,JIANGFan,LINXiang-hui,WULi-qiang.TheMotor-Drive ... 北京理工大学科研学术 本站小编 Free考研考试 2021-12-21
|