Visual object tracking using adaptive correlation filters
The DOI matches its Crossref record; 18 of the 19 references it cites were found in a registry or an index.
Abstract
Although not commonly used, correlation filters can track complex objects through rotations, occlusions and other distractions at over 20 times the rate of current state-of-the-art techniques. The oldest and simplest correlation filters use simple templates and generally fail when applied to tracking. More modern approaches such as ASEF and UMACE perform better, but their training needs are poorly suited to tracking. Visual tracking requires robust filters to be trained from a single frame and dynamically adapted as the appearance of the target object changes. This paper presents a new type of correlation filter, a Minimum Output Sum of Squared Error (MOSSE) filter, which produces stable correlation filters when initialized using a single frame. A tracker based upon MOSSE filters is robust to variations in lighting, scale, pose, and nonrigid deformations while operating at 669 frames per second. Occlusion is detected based upon the peak-to-sidelobe ratio, which enables the tracker to pause and resume where it left off when the object reappears.
Overview
Integrity signals
Registry
Cites
Statistics
Verification
Claims the paper makes
Citations
Cites
Breiman, L. (1996) · Machine Learning · doi:10.1007/bf00058655 · Cited by 16,325 · cited 1×
“Motivation for averaging comes from Bootstrap Aggregation [5] where the output of weak classifiers can be averaged to produce a much stronger classifier.”
BackgroundPress, W. H., Teukolsky, S. A., Vetterling, W. T., & Flannery, B. P. (1994) · Cited by 15,411
Incremental Learning for Robust Visual Tracking
Ross, D. A., Lim, J., Lin, R.-S., & Yang, M.-H. (2008) · International Journal of Computer Vision · doi:10.1007/s11263-007-0075-7 · Cited by 3,061 · cited 7×
“Recent examples include: Incremental Visual Tracking (IVT) [17], Robust Fragmentsbased Tracking (FragTrack) [1], Graph Based Discriminative Learning (GBDL) [19], and Multiple Instance Learning (MILTrack) [2].”
BackgroundMethodInfluentialVisual tracking with online Multiple Instance Learning
Babenko, B., Yang, M.-H., & Belongie, S. (2009) · 2009 IEEE Conference on Computer Vision and Pattern Recognition · doi:10.1109/cvpr.2009.5206737 · Cited by 1,770 · cited 6×
“Recent examples include: Incremental Visual Tracking (IVT) [17], Robust Fragmentsbased Tracking (FragTrack) [1], Graph Based Discriminative Learning (GBDL) [19], and Multiple Instance Learning (MILTrack) [2].”
BackgroundMethodInfluentialRobust Fragments-based Tracking using the Integral Histogram
Adam, A., Rivlin, E., & Shimshoni, I. (2006) · 2006 IEEE Computer Society Conference on Computer Vision and Pattern Recognition - Volume 1 (CVPR'06) · doi:10.1109/cvpr.2006.256 · Cited by 1,397 · cited 7×
“Recent examples include: Incremental Visual Tracking (IVT) [17], Robust Fragmentsbased Tracking (FragTrack) [1], Graph Based Discriminative Learning (GBDL) [19], and Multiple Instance Learning (MILTrack) [2].”
BackgroundMethodInfluentialRobust online appearance models for visual tracking
Jepson, A. D., Fleet, D. J., & El-Maraghi, T. F. (2003) · IEEE Transactions on Pattern Analysis and Machine Intelligence · doi:10.1109/tpami.2003.1233903 · Cited by 922
Minimum average correlation energy filters
Mahalanobis, A., Vijaya Kumar, B. V. K., & Casasent, D. (1987) · Applied Optics · doi:10.1364/ao.26.003633 · Cited by 598 · cited 1×
“In the 1980’s and 1990’s, many variants of correlation filters, including Synthetic Discriminant Functions (SDF) [7, 6], Minimum Variance Synthetic Discriminant Functions (MVSDF) [9], Minimum Average Correlation Energy (MACE) [11], Optimal Tradeoff Filters (OTF) [16], and Minimum Squared Error Synthetic Discriminant F…”
MethodMultivariant technique for multiclass pattern recognition
Hester, C. F., & Casasent, D. (1980) · Applied Optics · doi:10.1364/ao.19.001758 · Cited by 385 · cited 3×
“In [12], it was found that the hard constraints of SDF based filters like MACE caused issues with distortion Naive UMACE ASEF MOSSE INPUT FILTER OUTPUT Figure 2 .”
MethodUnconstrained correlation filters
Mahalanobis, A., Vijaya Kumar, B. V. K., Song, S., Sims, S. R. F., & Epperson, J. F. (1994) · Applied Optics · doi:10.1364/ao.33.003751 · Cited by 293 · cited 2×
“In fact, the optimization problem in Equation 3 is almost identical to optimization problems presented in [10] and [12].”
MethodUnified synthetic discriminant function computational formulation
Casasent, D. (1984) · Applied Optics · doi:10.1364/ao.23.001620 · Cited by 292 · cited 3×
“In [12], it was found that the hard constraints of SDF based filters like MACE caused issues with distortion Naive UMACE ASEF MOSSE INPUT FILTER OUTPUT Figure 2 .”
Method
Research works printed but not linked: 2
- messerschmitt, 2006, Stationary points of a real-valued function of a complex variable
- savvides, 2002, Face verification using correlation filters, AIAT
Cited by
Items 801–900 of 1,002
Hand tracking by extending distance transform and hand model in real-time
Kerdvibulvech, C. (2015) · Pattern Recognition and Image Analysis · doi:10.1134/s1054661815030098 · Cited by 18
DCCO: Towards Deformable Continuous Convolution Operators for Visual Tracking
Johnander, J., Danelljan, M., Khan, F. S., & Felsberg, M. (2017) · Lecture notes in computer science · doi:10.1007/978-3-319-64689-3_5 · Cited by 20
“[3] was restricted to a single feature channel and was later extended to multi-channel feature maps [11][10][12].”
BackgroundTouil, D. E., Terki, N., & Medouakh, S. (2017) · Applied Intelligence · doi:10.1007/s10489-017-1120-z · Cited by 18
“In addition, to remove the boundary discontinuities, each feature channel is weighted by a cosine window [25].”
BackgroundMethodOnline Multi-Object Tracking via Combining Discriminative Correlation Filters With Making Decision
Wu, C., Sun, H., Wang, H., Fu, K., Xu, G., Zhang, W., & Sun, X. (2018) · IEEE Access · doi:10.1109/access.2018.2858853 · Cited by 18
“Based on the work [4], a large number of investigations have been done to improve the accuracy and robustness of the tracking algorithm based on DCF.”
BackgroundMethodInfluentialBackground-Suppressed Correlation Filters for Visual Tracking
Chen, Z., Guo, Q., Wan, L., & Feng, W. (2018) · IEEE International Conference on Multimedia and Expo (ICME) · doi:10.1109/icme.2018.8486453 · Cited by 18
“tracking but beyond real-time speed [5, 6].”
Visual Tracking via Auto-Encoder Pair Correlation Filter
Cheng, X., Zhang, Y., Zhou, L., & Zheng, Y. (2019) · IEEE Transactions on Industrial Electronics · doi:10.1109/tie.2019.2913815 · Cited by 18
“the-art methods: MEEM [12], ACT [15], KCF [17], DSST [18], DeepSRDCF [22], ECO [23], SRDCF [25], SAMF [29], HCF [32], CCOT [26], SIT [35], SiameseFC [36], MDNet [37], and”
BackgroundMethodInfluentialThe Review of Target Tracking for UAV
Chen, P., & Zhou, Y. (2019) · IEEE Conference on Industrial Electronics and Applications (ICIEA) · doi:10.1109/iciea.2019.8833668 · Cited by 20
“[9] proposed a correlation filtering target tracking method based on Minimum Output Sum of Squared Error filter (MOSSE).”
MethodDing, M., Zhang, X., Chen, W., Wei, L., & Cao, Y. (2019) · Proceedings of the Institution of Mechanical Engineers Part G Journal of Aerospace Engineering · doi:10.1177/0954410019890820 · Cited by 19
Grayscale-Thermal Tracking via Inverse Sparse Representation-Based Collaborative Encoding
Kang, B., Liang, D., Ding, W., Zhou, H., & Zhu, W. (2019) · IEEE Transactions on Image Processing · doi:10.1109/tip.2019.2959912 · Cited by 21
“MOSSE [36]”
Double-Channel Object Tracking With Position Deviation Suppression
Chu, J., Tu, X., Leng, L., & Miao, J. (2019) · IEEE Access · doi:10.1109/access.2019.2961778 · Cited by 18
“The early correlation filter tracking algorithms use the low-level features [12]–[14] or their fusion [16].”
MethodMulti-Stream Siamese and Faster Region-Based Neural Network for Real-Time Object Tracking
Liu, Y., Zhang, L., Chen, Z., Yan, Y., & Wang, H. (2020) · IEEE Transactions on Intelligent Transportation Systems · doi:10.1109/tits.2020.3006927 · Cited by 19
Nousi, P., Triantafyllidou, D., Tefas, A., & Pitas, I. (2020) · Signal Processing Image Communication · doi:10.1016/j.image.2020.115969 · Cited by 20
Asymmetric discriminative correlation filters for visual tracking
Li, S., Jiang, Q.-B., Zhao, Q., Lu, L. P., & Feng, Z. (2020) · Frontiers of Information Technology & Electronic Engineering · doi:10.1631/fitee.1900507 · Cited by 19
“Using DCF for adaptive tracking was started with MOSSE by Bolme et al. (2010).”
MethodJaved, S., Mahmood, A., Dias, J. M. M., Seneviratne, L. D., & Werghi, N. (2021) · IEEE Transactions on Cybernetics · doi:10.1109/tcyb.2021.3086194 · Cited by 19
“In most of the DCF methods, an online correlation filter is trained from the region of interest in the current frame, which is then employed to track the target object in the subsequent frames by estimating the maximum response [3], [17].”
BackgroundMethodRank-Based Verification for Long-Term Face Tracking in Crowded Scenes
Barquero, G., Hupont, I., & Fernández, C. (2021) · IEEE Transactions on Biometrics Behavior and Identity Science · doi:10.1109/tbiom.2021.3099568 · Cited by 18
“Other highperforming visual trackers include: MOSSE [23] and CSRT [24], which are both based on correlation filters; and the Median Flow”
MethodInfluentialRAMC: A Rotation Adaptive Tracker with Motion Constraint for Satellite Video Single-Object Tracking
Chen, Y., Tang, Y., Han, T., Zhang, Y., Zou, B., & Feng, H. (2022) · Remote Sensing · doi:10.3390/rs14133108 · Cited by 18
“Correlation filter-based methods have emerged since MOSSE [21] was first proposed.”
MethodRGB-T object tracking via sparse response-consistency discriminative correlation filters
Huang, Y., Li, X., Lü, R., & Qi, N. (2022) · Infrared Physics & Technology · doi:10.1016/j.infrared.2022.104509 · Cited by 19
Glitzner, M., Crijns, S. P. M., de Senneville, B. D., Lagendijk, J. J. W., & Raaymakers, B. W. (2015) · Physics in Medicine and Biology · doi:10.1088/0031-9155/60/5/2005 · Cited by 18
“The processing module extracts the pixel location of the marker based on a fast in-house minimum output sum of squared error (MOSSE) filter implementation (Bolme et al. 2010, Crijns, Kotte, Lagendijk & Raaymakers 2012), tuned for working in a 64 × 64px local region.”
MethodSpatial pyramid context-aware moving vehicle detection and tracking in urban aerial imagery
Poostchi, M., Palaniappan, K., & Seetharaman, G. S. (2017) · IEEE International Conference on Advanced Video and Signal Based Surveillance (AVSS) · doi:10.1109/avss.2017.8078504 · Cited by 18
“Table 1 presents the computed measures on 20 sequences from ARGUS WAMI dataset for SPCT and 9 trackers including correlation-based trackers SWCF [8], KCF [13], Staple [7], SAMF [17] and NSAMF [20], optical flow clustering tracker FCT [15], DPCF [3] and normalized crosscorrelation tracker NCC [17].”
MethodMulti-Channel Features Spatio-Temporal Context Learning for Visual Tracking
Zhou, X., Liu, X., Yang, C., Jiang, A., & Yan, B. (2017) · IEEE Access · doi:10.1109/access.2017.2720746 · Cited by 17
“As described in the convolution theorem [11], the correlation in the time domain corresponds to an element-wise multiplication in the Fourier domain.”
MethodA Reliable and Real-Time Tracking Method with Color Distribution
Zhao, Z., Han, Y., Xu, T., Li, X., Song, H., & Luo, J. (2017) · Sensors · doi:10.3390/s17102303 · Cited by 17
“introduced the correlation filter into tracking [14].”
Porfiriev, A. V., Сумин, В. И., & Dushkin, A. V. (2017) · doi:10.1109/uralcon.2017.8120710 · Cited by 17
“In the process of learning uses the method of moving average [14, 23, 24]:”
MethodRobust Visual Tracking via an Improved Background Aware Correlation Filter
Sheng, X., Liu, Y., Liang, H., Li, F., & Man, Y. (2019) · IEEE Access · doi:10.1109/access.2019.2900666 · Cited by 17
“, [7]–[13], [24], [26], [27], [31] and references therein.”
MethodParfiryev, A. V., Ischuk, I. N., Dushkin, A. V., С., Б. Т., & Popova, N. A. (2019) · IEEE Conference of Russian Young Researchers in Electrical and Electronic Engineering · doi:10.1109/eiconrus.2019.8656636 · Cited by 17
“As a result of standard transformations, the final appearance of the filter is determined [14]:”
BackgroundFast-deepKCF Without Boundary Effect
Zheng, L., Tang, M., Chen, Y., Wang, J., & Lu, H. (2019) · IEEE/CVF International Conference on Computer Vision (ICCV) · doi:10.1109/iccv.2019.00412 · Cited by 19
“Since MOSSE [3], almost all CF trackers [22, 39, 18, 28, 10, 1, 12, 13, 14, 24, 8, 31] have been relying on fast Fourier transform (FFT) to accelerate their computations.”
BackgroundMethodVisible–Infrared Dual-Sensor Fusion for Single-Object Tracking
Liu, W., Liu, W., & Sun, Y. (2023) · IEEE Sensors Journal · doi:10.1109/jsen.2023.3234091 · Cited by 18
Monocular-Vision-Based Moving Target Geolocation Using Unmanned Aerial Vehicle
Pan, T., Deng, B., Dong, H., Gui, J., & Zhao, B. (2023) · Drones · doi:10.3390/drones7020087 · Cited by 17
Repformer: a robust shared-encoder dual-pipeline transformer for visual tracking
Gu, F., Lu, J., Cai, C., Zhu, Q., & Ju, Z. (2023) · Neural Computing and Applications · doi:10.1007/s00521-023-08824-2 · Cited by 17
“In the correlation filtering-based tracking methods [29], the primary concept is to evaluate the correlation [13, 14] between the template image and the input video image.”
BackgroundMethodInfluentialIoUformer: Pseudo-IoU prediction with transformer for visual tracking
Cai, H., Lan, L., Zhang, J., Zhang, X., Zhan, Y., & Luo, Z. (2023) · Neural Networks · doi:10.1016/j.neunet.2023.10.055 · Cited by 18
RTSformer: A Robust Toroidal Transformer With Spatiotemporal Features for Visual Tracking
Gu, F., Lu, J., Cai, C., Zhu, Q., & Ju, Z. (2024) · IEEE Transactions on Human-Machine Systems · doi:10.1109/thms.2024.3370582 · Cited by 17
Qian, K., Zhu, D., Wu, Y., Shen, J., & Zhang, S. (2024) · Infrared Physics & Technology · doi:10.1016/j.infrared.2024.105674 · Cited by 18
Jenkins, M. D., Barrie, P., Buggy, T., & Morison, G. (2015) · Pattern Recognition Letters · doi:10.1016/j.patrec.2015.10.014 · Cited by 17
Context-based occlusion detection for robust visual tracking
Niu, X., & Qiao, Y. (2017) · Proceedings - International Conference on Image Processing · doi:10.1109/icip.2017.8296964 · Cited by 18
“[1] proposed Minimum Output Sum of Squared Error (MOSSE), where the appearance of target is modeled by correlation filter and by correlating the trained filter with candidates of target, the position of target in the current frame is the position which has the strongest response.”
BackgroundScale Adaptive Kernelized Correlation Filter Tracker with Feature Fusion
Zhou, T., Zhu, M. Q., Zeng, D., & Yang, H. (2017) · Mathematical Problems in Engineering · doi:10.1155/2017/1605959 · Cited by 18
“Third, learning rate is fixed inmost of the existing correlation filters [14].”
BackgroundHard negative mining for correlation filters in visual tracking
Sun, Z., Wang, Y., & Laganière, R. (2019) · Machine Vision and Applications · doi:10.1007/s00138-019-01004-0 · Cited by 17
“The pioneering work of (MOSSE) [2] ignited the interest”
BackgroundInfluentialDecision Controller for Object Tracking With Deep Reinforcement Learning
Zhong, Z., Yang, Z., Feng, W., Wu, W., Hu, Y., & Liu, C. (2019) · IEEE Access · doi:10.1109/access.2019.2900476 · Cited by 17
“Most of these methods [10]–[13] use handcrafted features and some [14]–[16] tend to combination of DCF framework and CNN features.”
BackgroundMethodVisual Tracking by Structurally Optimizing Pre-Trained CNN
Liu, C., Liu, P., Zhao, W., & Tang, X. (2019) · IEEE Transactions on Circuits and Systems for Video Technology · doi:10.1109/tcsvt.2019.2938038 · Cited by 19
Contour-Aware Long-Term Tracking With Reliable Re-Detection
Tang, F., & Ling, Q. (2019) · IEEE Transactions on Circuits and Systems for Video Technology · doi:10.1109/tcsvt.2019.2957748 · Cited by 18
“For hand-crafted feature based correlation filter tracking, the peak-to-sidelobe ratio (PSR) is computed to quantify the sharpness of the response peak [16].”
MethodPoint Siamese Network for Person Tracking Using 3D Point Clouds
Cui, Y., Fang, Z., & Zhou, S. (2019) · Sensors · doi:10.3390/s20010143 · Cited by 23
“[18] added HOG features based on [17] and has a better performance.”
BackgroundContext-aware MDNet for target tracking in UAV remote sensing videos
Bi, F., Lei, M., & Wang, Y. (2020) · International Journal of Remote Sensing · doi:10.1080/01431161.2019.1708500 · Cited by 17
“To select the best tracker after embedding the RA-CACF in our network, we compare four CF trackers, including SAMF (Li and Zhu 2014), STAPLE (Bertinetto et al. 2016a), KCF (Henriques et al. 2015), and MOSSE (Bolme et al. 2010), using UAV123, as shown in Table 1.”
MethodUAV object tracking by background cues and aberrances response suppression mechanism
Tian, L., Ding, F., & Yang, W. (2020) · Neural Computing and Applications · doi:10.1007/s00521-020-05200-2 · Cited by 17
“The original CF tracker [3] was proposed in 2010 and achieved a tracking speed of more than 600 frame-per-second (FPS), which attracted widespread attention.”
MethodSpatio-temporal joint aberrance suppressed correlation filter for visual tracking
Xu, L., Kim, P., Wang, M., Pan, J., Yang, X., & Gao, M. (2021) · Complex & Intelligent Systems · doi:10.1007/s40747-021-00544-1 · Cited by 17
“[5] first proposed the correlation filter tracker, termed minimum output sum of squared error (MOSSE), to learn a filter between multiple training image patches and a template of user-specified ideal correlation response.”
MethodAuto-Perceiving Correlation Filter for UAV Tracking
Wang, L., Li, J., Huang, B., Chen, J., Li, X., Wang, J., & Xu, T. (2022) · IEEE Transactions on Circuits and Systems for Video Technology · doi:10.1109/tcsvt.2022.3155731 · Cited by 18
Tang, W., Han, K., Zhang, H., Yu, P., Arnold, C. W., & Zhang, R. (2022) · Lecture notes in computer science · doi:10.1007/978-3-031-16446-0_19 · Cited by 21
“In the field of natural images, there is a similar problem called target tracking or object tracking, for which several deep learning-based methods have been proposed [3,12,23].”
Toward Autonomous Localization of Planetary Robotic Explorers by Relying on Semantic Mapping
Ebadi, K., Coble, K., Kogan, D., Atha, D., Schwartz, R., Padgett, C. W., & Hook, J. V. (2022) · IEEE Aerospace Conference (AERO) · doi:10.1109/aero53065.2022.9843550 · Cited by 17
“The method adapts the MOSSE correlation filter [29] to determine a position estimate of the query frame with GPS-level accuracy by rendering a view based on DEM data from a known camera heading and FOV at each point in an x-y grid of points surrounding the true location of the vessel.”
MethodSpatio-temporal matching for siamese visual tracking
Zhang, J., Dai, K., Li, Z., Wei, R., & Wang, Y. (2022) · Neurocomputing · doi:10.1016/j.neucom.2022.11.093 · Cited by 17
Scalable Decision-Theoretic Coordination and Control for Real-time Active Multi-Camera Surveillance
Natarajan, P., Hoang, T. N., Wong, Y., Low, K. H., & Kankanhalli, M. (2014) · International Conference on Distributed Smart Cameras · doi:10.1145/2659021.2659042 · Cited by 17
“We employed a correlation filters based tracker [2] which reported to operate at 669 frames per second on single target.”
MethodNon-parametric Bayesian Constrained Local Models
Martins, P., Caseiro, R., & Batista, J. P. (2014) · IEEE Computer Society Conference on Computer Vision and Pattern Recognition/Proceedings - IEEE Computer Society Conference on Computer Vision and Pattern Recognition/Proceedings · doi:10.1109/cvpr.2014.232 · Cited by 18
“The desired MOSSE correlation output (see [3]) was set to be a 2D Gaussian centered at the each landmark with 3 pixels of standard deviation.”
BackgroundMethodInfluentialMoving object detection and tracking using deep learning neural network and correlation filter
Supreeth, H. S. G., & Patil, C. M. (2018) · International Conference Inventive Communication and Computational Technologies · doi:10.1109/icicct.2018.8473354 · Cited by 19
“Therefore, the filters are required to adapt to these dynamic changes for tracking the object efficiently purpose [35].”
BackgroundControl of unmanned aerial vehicles based on the detection algorithm
Parfiryev, A. V., Dushkin, A. V., Дубровин, А. С., & Stepanov, L. V. (2019) · Journal of Physics Conference Series · doi:10.1088/1742-6596/1202/1/012014 · Cited by 17
Vehicle tracking based on shape information and inter-frame motion vector
Liu, P., Wang, G., Yu, Z., Guo, X., & Lu, W. (2019) · Computers & Electrical Engineering · doi:10.1016/j.compeleceng.2019.06.019 · Cited by 18
Release the Power of Online-Training for Robust Visual Tracking
Yang, Y., Li, G., Qi, Y., & Huang, Q. (2020) · Proceedings of the AAAI Conference on Artificial Intelligence · doi:10.1609/aaai.v34i07.6956 · Cited by 19
“Bolme etc. (Bolme et al. 2010) propose a method to reject new samples based on the Peak-to-Sidelobe Ratio (PSR) criterion.”
MethodBoosted Kernelized Correlation Filters for Event-based Face Detection
Ramesh, B., & Yang, H. (2020) · IEEE/CVF Winter Conference on Applications of Computer Vision Workshops (WACVW) · doi:10.1109/wacvw50321.2020.9096944 · Cited by 19
“As mentioned in [1], the regularization parameter alleviates the problem of zero-frequency components in the spectrum of f , which would lead to division by zero.”
BackgroundSiamese Keypoint Prediction Network for Visual Object TrackingPreprint
Li, Q., Qin, Z., Zhang, W., & Zheng, W. (2020) · arXiv (Cornell University) · doi:10.48550/arxiv.2006.04078 · Cited by 22
“In literature, a classical approach to visual object tracking is the discriminative correlation filter [5,18].”
Tracking Grow-Finish Pigs Across Large Pens Using Multiple CamerasPreprint
Shirke, A., Saifuddin, A., Luthra, A., Li, J., Williams-Stroud, T. N., Hu, X., Kotnana, A., Kocabalkanli, O., Ahuja, N., Green-Miller, A. R., et al. (2021) · arXiv (Cornell University) · doi:10.48550/arxiv.2111.10971 · Cited by 27
“[21] uses an SSD [9] architecture coupled with the MOSSE [4] algorithm to perform animal tracking.”
Method2D visual tracking for sports UAV cinematography applications
Zachariadis, O., Mygdalis, V., Mademlis, I., Nikolaidis, N., & Pitas, I. (2017) · IEEE Global Conference on Signal and Information Processing · doi:10.1109/globalsip.2017.8308599 · Cited by 16
“As have been proposed by [12], the Minimum Output Sum of Squared Error (MOSSE) filter is updated as follows:”
MethodFan, J., Song, H., Zhang, K., Liu, Q., & Lian, W. (2018) · IEEE Access · doi:10.1109/access.2018.2872691 · Cited by 16
“UsingDCFs for visual tracking starts withMOSSE [6], which learns CFs with a few samples in the frequency domain, thereby facilitating fast Fourier transforms (FFTs) for efficient computation that runs at 669 frames per second (FPS).”
BackgroundReal-Time Visual Tracking with Compact Shape and Color Feature
Gao, Z., Xia, S., Zhang, Y., Yao, R., Zhao, J., Niu, Q., & Jiang, H. (2018) · Computers, materials & continua/Computers, materials & continua (Print) · doi:10.3970/cmc.2018.02634 · Cited by 17
Shan, Y., Liu, S., Zhang, Y., Jing, M., & Xu, H. (2021) · IEEE Access · doi:10.1109/access.2021.3079132 · Cited by 16
“Based on MOSSE, CSK [26] introduced the kernel method and dense sampling single-channel gray features.”
MethodZhang, F., Ma, S., Qiu, Z.-L., & Qi, T. (2021) · Signal Processing · doi:10.1016/j.sigpro.2021.108352 · Cited by 16
Robust Online Tracking With Meta-Updater
Zhao, J. W., Dai, K., Zhang, P., Wang, D., & Lu, H. (2022) · IEEE Transactions on Pattern Analysis and Machine Intelligence · doi:10.1109/tpami.2022.3202785 · Cited by 19
DeepPCT: Single Object Tracking in Dynamic Point Cloud Sequences
Liu, H., Hu, Q., Ma, Y., Xie, K., & Guo, Y. (2022) · IEEE Transactions on Instrumentation and Measurement · doi:10.1109/tim.2022.3232092 · Cited by 21
“1) 2-D SOT: Early works in this area mainly focus on correlation filters [10], [11], while most current advanced methods are based on CNNs and Siamese networks [12].”
BackgroundWang, S., Qian, K., Shen, J., Ma, H., & Chen, P. (2023) · Remote Sensing · doi:10.3390/rs15071731 · Cited by 17
“The pioneer Minimum Output Sum of Squared Error (MOSSE) algorithm [14] adopts the gray-scale feature to complete target tracking with high execution efficiency.”
MethodZhang, H., Liu, G., Zhang, Y., & Hao, Z. (2023) · IEEE Transactions on Multimedia · doi:10.1109/tmm.2023.3289700 · Cited by 16
Visualizing Skiers' Trajectories in Monocular Videos
Dunnhofer, M., Sordi, L., & Micheloni, C. (2023) · IEEE/CVF Conference on Computer Vision and Pattern Recognition Workshops (CVPRW) · doi:10.1109/cvprw59228.2023.00547 · Cited by 17
“For the visual tracker, we evaluated: the deep learning-based methodologies STARK [67], SuperDiMP [2, 32], SiamRPN++ [36], trained for generic object tracking; and the traditional method MOSSE [3].”
MethodAn Adaptive Correlation Filtering Method for Text-Based Person Search
Sun, M., Suo, W., Wang, P., Niu, K., Liu, L., Lin, G., Zhang, Y., & Wu, Q. (2024) · International Journal of Computer Vision · doi:10.1007/s11263-024-02094-8 · Cited by 17
Ntousis, O., Makris, E., Tsanakas, P. D., & Pavlatos, C. (2025) · Technologies · doi:10.3390/technologies13010035 · Cited by 16
Yu, Y., Chen, Z., Zhang, Y., Zhang, C., & Ding, W. (2025) · IEEE Transactions on Intelligent Transportation Systems · doi:10.1109/tits.2025.3533953 · Cited by 17
“The minimum output sum of squared errors (MOSSE) [40] tracker is a pioneering method based on correlation filters (CFs).”
Becker, S., Krah, S. B., Hübner, W., & Arens, M. (2016) · Optics and Photonics for Counterterrorism, Crime Fighting, and Defence · doi:10.1117/12.2243473 · Cited by 16
Learning adaptively windowed correlation filters for robust tracking
Kuai, Y., Wen, G., & Li, D. (2018) · Journal of Visual Communication and Image Representation · doi:10.1016/j.jvcir.2018.01.008 · Cited by 16
A stable long-term object tracking method with re-detection strategy
Li, T., Zhao, S., Meng, Q., Chen, Y., & Shen, J. (2018) · Pattern Recognition Letters · doi:10.1016/j.patrec.2018.09.017 · Cited by 17
Learning target-aware correlation filters for visual tracking
Li, D., Wen, G., Kuai, Y., Xiao, J., & Porikli, F. (2018) · Journal of Visual Communication and Image Representation · doi:10.1016/j.jvcir.2018.11.036 · Cited by 16
“The pioneer MOSSE tracker [18] achieves an impressive tracking speed of over 600 fps.”
BackgroundDSNet: Deep and Shallow Feature Learning for Efficient Visual Tracking
Wu, Q., Yan, Y., Liang, Y., Liu, Y., & Wang, H. (2019) · Lecture notes in computer science · doi:10.1007/978-3-030-20873-8_8 · Cited by 16
Context constraint and pattern memory for long-term correlation tracking
Huang, B., Xu, T., Лю, Б., & Yuan, B. (2019) · Neurocomputing · doi:10.1016/j.neucom.2019.10.021 · Cited by 16
Multiple object tracking in soccer videos using topographic surface analysis
Kim, W. (2019) · Journal of Visual Communication and Image Representation · doi:10.1016/j.jvcir.2019.102683 · Cited by 18
Benchmarking Deep Trackers on Aerial Videos
Taufique, A. M. N., Minnehan, B., & Savakis, A. E. (2020) · Sensors · doi:10.3390/s20020547 · Cited by 16
Zhang, D., Lv, J., & Cheng, Z. (2020) · IEEE Access · doi:10.1109/access.2020.3003229 · Cited by 17
“The vehicle tracking results are tested using multiple datasets, such as the LaSOT (Large-scale Single Object Tracking), VOT2017 (2017 Visual Object Tracking), and OTB2015 (2015 Object Tracking Benchmark) datasets, and various algorithms, such as the DLT, MOSSE (minimum output sum of squared error) filter, DCF (discrim…”
MethodIntegration of the 3D Environment for UAV Onboard Visual Object Tracking
Vujasinović, S., Becker, S., Breuer, T., Bullinger, S., Scherer‐Negenborn, N., & Arens, M. (2020) · Applied Sciences · doi:10.3390/app10217622 · Cited by 19
“Currently, three tracking designs prevail on those benchmarks: the discriminative correlation filters [14–20]; the Siamese-based approach [21–26]; and recently, trackers inspired by correlation filters that employ a small convolutional neural network for learning the appearance model of the object [27,28].”
BackgroundTowards Robust Visual Tracking for Unmanned Aerial Vehicle with Tri-Attentional Correlation Filters
He, Y., Fu, C., Lin, F., Li, Y., & Lu, P. (2020) · IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS) · doi:10.1109/iros45743.2020.9341784 · Cited by 18
“s per second (FPS). II. RELATED WORKS A. Tracking with correlation filters Exploiting CF for object tracking started with the method called the minimum output sum of squared error, i.e., MOSSE tracker [18]. The tracker is constructed and trained using gray-scale samples in the frequency domain for efficiency. Afterward,…”
MethodTICNet: A Target-Insight Correlation Network for Object Tracking
Ruan, W., Ye, M., Wu, Y., Liu, W., Chen, J., Liang, C., Li, G., & Lin, C. (2021) · IEEE Transactions on Cybernetics · doi:10.1109/tcyb.2021.3070677 · Cited by 16
“Inspired by the pioneering work MOSSE [9], CF trackers have been studied”
MethodInfluentialTrapal, D. D. C., Leong, B. C. C., Ng, H. W., Zhong, J. T. G., Srigrarom, S. S., & Chan, T. H. (2021) · International Conference on Control and Robotics Engineering (ICCRE) · doi:10.1109/iccre51898.2021.9435671 · Cited by 17
“The DCF tracker tracks an object by using discriminative correlation methods where a filter is learnt based on the training image [14].”
MethodOptical Tracking System for Multi-UAV Clustering
Ming, R., Zhou, Z., Luo, X., Liu, W., Zi, L., Song, C., Jiang, R., & Zang, Y. (2021) · IEEE Sensors Journal · doi:10.1109/jsen.2021.3091280 · Cited by 16
Correlation filters with adaptive convolution response fusion for object tracking
Yang, L., Kong, C., Chang, X., Zhao, S., Cao, Y., & Zhang, S. (2021) · Knowledge-Based Systems · doi:10.1016/j.knosys.2021.107314 · Cited by 18
Deep learning assisted visual tracking of evader-UAV
Τσουκαλάς, Α., Xing, D., Evangeliou, N., Giakoumidis, N., & Tzes, A. P. (2021) · International Conference on Unmanned Aircraft Systems (ICUAS) · doi:10.1109/icuas51884.2021.9476720 · Cited by 16
Dynamic Particle Filter Framework for Robust Object Tracking
Li, S., Zhao, S., Cheng, B. H., & Chen, J. (2021) · IEEE Transactions on Circuits and Systems for Video Technology · doi:10.1109/tcsvt.2021.3109981 · Cited by 17
Efficient Online Object Tracking Scheme for Challenging Scenarios
Mehmood, K., Ali, A., Jalil, A., Khan, B., Cheema, K. M., Murad, M., & Milyani, A. H. (2021) · Sensors · doi:10.3390/s21248481 · Cited by 16
“MOSSECA is an improved context-aware formulation version of the MOSSE [64] tracker.”
BackgroundSiamese block attention network for online update object tracking
Xiao, D., Tan, K., Wei, Z., & Zhang, G. (2022) · Applied Intelligence · doi:10.1007/s10489-022-03619-9 · Cited by 18
“MOOSE[15] first applies the correlation filter in the visual tracking task and shows the potential of correlation-filter based trackers.”
MethodZhu, S., McCullough, M., Pujic, Z., Sibberas, J., Sun, B., Darveniza, T., Bucknall, B., Avitan, L., & Goodhill, G. J. (2023) · Journal of Neuroscience · doi:10.1523/jneurosci.1721-22.2022 · Cited by 16
Yang, J., & Tse, P. W.-T. (2023) · Measurement · doi:10.1016/j.measurement.2023.112663 · Cited by 16
Assessment of experimental OpenCV tracking algorithms for ultrasound videos
Levin, A. A., Klimov, D. D., Nechunaev, A., Prokhorenko, L. S., Mishchenkov, D., Nosova, A. G., Астахов, Д. А., Poduraev, Y. V., & Панченков, Д. Н. (2023) · Scientific Reports · doi:10.1038/s41598-023-30930-3 · Cited by 16
Slight Aware Enhancement Transformer and Multiple Matching Network for Real-Time UAV Tracking
Deng, A., Han, G., Chen, D., Ma, T., & Liu, Z. (2023) · Remote Sensing · doi:10.3390/rs15112857 · Cited by 17
“Current advanced tracking methods can be divided into two directions: correlation filters [4] and siamese networks [5].”
BackgroundSiamIST: Infrared small target tracking based on an improved SiamRPN
Qian, K., Zhang, S.-J., Ma, H., & Sun, W. (2023) · Infrared Physics & Technology · doi:10.1016/j.infrared.2023.104920 · Cited by 17
Multiple templates transformer for visual object tracking
Pang, H., Su, J., Ma, R., Li, T., & Liu, C. (2023) · Knowledge-Based Systems · doi:10.1016/j.knosys.2023.111025 · Cited by 17
DeePLS: Personalize Lighting in Smart Home by Human Detection, Recognition, and Tracking
Sobhani, A., Khorshidi, F., & Fakhredanesh, M. (2023) · SN Computer Science · doi:10.1007/s42979-023-02240-y · Cited by 17
“Their work is built on the MOSSE tracker [31].”
BackgroundInfluentialGenerating Reliable Online Adaptive Templates for Visual Tracking
Guo, J., Xu, T., Jiang, S., & Shen, Z. (2018) · Proceedings - International Conference on Image Processing · doi:10.1109/icip.2018.8451440 · Cited by 18
“The correlation filter based trackers [7,8] use linear interpolation to update their filters after each frame.”
Understanding Tracking Methodology of Kernelized Correlation Filter
Yadav, S., & Payandeh, S. (2018) · IEEE Annual Information Technology, Electronics and Mobile Communication Conference (IEMCON) · doi:10.1109/iemcon.2018.8614990 · Cited by 19
“[20], correlation based filters have enjoyed immense popularity in tracking community.”
BackgroundA Visual Leader-Following Approach With a T-D-R Framework for Quadruped Robots
Pang, L., Cao, Z., Yu, J., Guan, P., Rong, X., & Chai, H. (2019) · IEEE Transactions on Systems Man and Cybernetics Systems · doi:10.1109/tsmc.2019.2912715 · Cited by 19
“In (3), y is the desired correlation output (training labels), which is typically selected to be Gaussian function with a parametrized standard deviation [31].”
BackgroundQu, Z., Yi, W., Zhou, R., Wang, H., & Chi, R. (2019) · IEEE Access · doi:10.1109/access.2019.2939435 · Cited by 16
“For the latter, the MOSSE [10] correlation filter was developed by David S.”
MethodiPAC: Integrate Pedestrian Dead Reckoning and Computer Vision for Indoor Localization and Tracking
Li, D., Lu, Y., Xu, J., Ma, Q., & Liu, Z. (2019) · IEEE Access · doi:10.1109/access.2019.2960287 · Cited by 16
“At present, the visual tracking method is mainly divided into two categories: deep learning [18], [19] and correlation filter [17], [20], [21].”
MethodMultiview Supervision By Registration
Zhang, Y., & Park, H. S. (2020) · IEEE Winter Conference on Applications of Computer Vision · doi:10.1109/wacv45572.2020.9093591 · Cited by 19
“Temporal Supervision The tracking results such as optical flow [3], MOSSE [5], and discriminative correlation filters [13], provides an auxiliary information that can be used to enforce the temporal consistency across a continuous sequence [8, 35].”
Metadata
- Authors
- Bolme, D., Beveridge, J. R., Draper, B. A., & Lui, Y. M.
- Affiliations
- Colorado State University (US)
- Venue
- 2010 IEEE Computer Society Conference on Computer Vision and Pattern Recognition
- Publisher
- IEEE
- Published
- 2010-06
- Pages
- 2544-2550
- ISSN
- 1063-6919 (ISSN-L)
- OpenAlex
- W1964846093
- Cited by
- 3,406
- References
- 19
- Citations by year
- 2026: 78 · 2025: 165 · 2024: 246 · 2023: 327 · 2022: 391 · 2021: 442 · 2020: 433 · 2019: 493 · 2018: 372 · 2017: 206 · 2016: 118 · 2015: 71 · 2014: 36 · 2013: 16 · 2012: 8
- FWCI
- 7.68
- Type
- conference paper
- Language
- en
- Open access
- closed
- Locations
- IEEE Computer Society Conference on Computer Vision and Pattern Recognition/Proceedings - IEEE Computer Society Conference on Computer Vision and Pattern Recognition/Proceedings (publishedVersion)
- Topics
- Physical Sciences › Computer Science › Computer Vision and Pattern Recognition › Video Surveillance and Tracking Methods
- Physical Sciences › Computer Science › Computer Vision and Pattern Recognition › Advanced Image and Video Retrieval Techniques
- Physical Sciences › Computer Science › Computer Vision and Pattern Recognition › Infrared Target Detection Methodologies
- Keywords
- visual object tracking, correlation filter, occlusion detection, peak-to-sidelobe ratio, non-rigid deformation, real-time tracking
About these checks
Registry, retraction, Cites and statistics results are computed, not judged, with one exception: a printed reference that no index lists is read once by a model before it counts as a citation error. Claims are extracted by a model and kept only when their quote appears verbatim in the text; each cited source is then checked, with a model’s help, for whether it says what the claim relies on. Those verdicts can be wrong — read the claim’s quote and the cited source before relying on one.
Statistics and arithmetic are recomputed, not judged.
Confidence bands are shown only for fields calibrated on our annotated test set. None is calibrated yet, so no band is shown.
What counts as a claim ›How we measure confidence ›How we check this ›
Are you an author, or do you see a problem with this page? Request a review
Data sources
Bibliographic data: Semantic Scholar, OpenAlex, Crossref.
Contains information from the Semantic Scholar Academic Graph Datasets, made available under the ODC Attribution License (ODC-BY 1.0).
Semantic Scholar’s data is described in: Kinney, R., et al. (2023). The Semantic Scholar Open Data Platform. arXiv:2301.10140.