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 701–800 of 1,002
Small Target Recognition and Tracking Based on UAV Platform
Tian, X., Jia, Y., Luo, X., & Yin, J. (2022) · Sensors · doi:10.3390/s22176579 · Cited by 22
“After the first correlation filtering method, Minimum Output Sum of Squared Error (MOOSE) [23], a large number of methods based on correlation filtering have been proposed one after another.”
MethodLong-term Visual Tracking: Review and Experimental Comparison
Liu, C., Chen, X., Bo, C., & Wang, D. (2022) · Machine Intelligence Research · doi:10.1007/s11633-022-1344-1 · Cited by 24
Zhang, Z., Zhu, X., Zhao, D., Arun, P. V., Zhou, H., Qian, K., & Hu, J. (2022) · Remote Sensing · doi:10.3390/rs14235958 · Cited by 22
“In addition, many other features are also used to improve the tracking robustness, including HOG [24], color name features [25], and convolution features [26].”
MethodInfluentialCMT: Context-Matching-Guided Transformer for 3D Tracking in Point Clouds
Guo, Z., Mao, Y., Zhou, W., Wang, M., & Li, H. (2022) · Lecture notes in computer science · doi:10.1007/978-3-031-20047-2_6 · Cited by 34
“Within these approaches, the Siamese tracking paradigm has demonstrated its advantages over the traditional discriminative correlation filter [4,20,12,21,11].”
MethodDomain Adaptive Transformer Tracking Under Occlusions
Yu, Q., Fan, K., & Zheng, Y. (2023) · IEEE Transactions on Multimedia · doi:10.1109/tmm.2023.3234372 · Cited by 25
“Traditional tracking methods [1], [2], [3], [4] use handcrafted features to describe the representations of objects and train discriminative or generative models to achieve tracking.”
MethodPortalés, C., Gimeno, J., Salvador, A., García-Fadrique, A., & Casas, S. (2022) · Computers & Graphics · doi:10.1016/j.cag.2022.12.006 · Cited by 22
Zhang, Y., Yu, Y., Chen, L., & Ding, W. (2023) · IEEE Transactions on Geoscience and Remote Sensing · doi:10.1109/tgrs.2023.3325337 · Cited by 26
“Minimum output sum of squared error (MOSSE) filter [24] is one of the pioneers who uses the least squares method to compute filter parameters.”
MethodExploiting Spatial Structure from Parts for Adaptive Kernelized Correlation Filter Tracker
Yao, R., Xia, S., Shen, F., Zhou, Y. Z., & Niu, Q. (2016) · IEEE Signal Processing Letters · doi:10.1109/lsp.2016.2545705 · Cited by 22
“Correlation filters are able to transfer the operator into the Fourier domain as element-wise multiplication readily, it is computationally efficiency.”
Discriminative Bayesian Active Shape Models
Martins, P., Caseiro, R., Henriques, J. F., & Batista, J. P. (2012) · Lecture notes in computer science · doi:10.1007/978-3-642-33712-3_5 · Cited by 22
“The Minimum Output Sum of Squared Error (MOSSE) filter, recently proposed in [11], finds the optimal filter that minimizes the Sum of Squared Differences (SSD) to a desired correlation output.”
BackgroundMethodInfluentialIdentification of ASD Children based on Video Data
Li, J., Zhong, Y., & Ouyang, G. (2018) · Proceedings - International Conference on Pattern Recognition/Proceedings/International Conference on Pattern Recognition · doi:10.1109/icpr.2018.8545113 · Cited by 25
“Correlation tracking methods such as kernelized correlation filters (KCF) [23] and the Minimum Output Sum of Squared Error (MOSSE) tracker [24] are based on correlation filters.”
MethodReliable correlation tracking via dual-memory selection model
Li, G., Peng, M., Nai, K., Li, Z., & Li, K. (2020) · Information Sciences · doi:10.1016/j.ins.2020.01.015 · Cited by 22
Siamese Deformable Cross-Correlation Network for Real-Time Visual Tracking
Zheng, L., Chen, Y., Tang, M., Wang, J., & Lu, H. (2020) · Neurocomputing · doi:10.1016/j.neucom.2020.02.080 · Cited by 23
Gaze Control of a Robotic Head for Realistic Interaction With Humans
Domingo, J. D., Gómez‐García‐Bermejo, J., & Zalama, E. (2020) · Frontiers in Neurorobotics · doi:10.3389/fnbot.2020.00034 · Cited by 23
Harikrishnan, P. M., Thomas, A., Gopi, V. P., Palanisamy, P., & Wahid, K. A. (2021) · Applied Intelligence · doi:10.1007/s10489-020-02127-y · Cited by 22
“In terms of both speed and mean precision, KCF outperforms other common object tracking algorithms such as Tracking-Learning-Detection (TLD) [40], Structured Output Tracking with Kernels (STRUCK) [41], Minimum Output Sum of Squared Error (MOSSE) [42], etc.”
MethodCorrelation filter‐based visual tracking via adaptive weighted CNN features fusion
Hao, Z., Liu, G., & Zhang, H. (2018) · IET Image Processing · doi:10.1049/iet-ipr.2017.0443 · Cited by 22
Cui, Z., Cai, Y., Zheng, W., Xu, C., & Yang, J. J. (2018) · IEEE Transactions on Image Processing · doi:10.1109/tip.2018.2886788 · Cited by 24
Airborne optical imaging technology: a road map in CIOMP
Jia, P., Tian, D., Wang, Y., Yao, D., Xu, R. L., Sun, J., Sun, H. G., & Zhang, B. (2025) · Light Science & Applications · doi:10.1038/s41377-025-01776-3 · Cited by 25
Zhu, G., Wang, J., Wang, P., Wu, Y., & Lu, H. (2017) · IEEE Transactions on Cybernetics · doi:10.1109/tcyb.2017.2776977 · Cited by 21
“by minimizing the output error over all the historical tracked results [17], [19].”
Surrounding-aware correlation filter for UAV tracking with selective spatial regularization
Fu, C., Xiong, W., Lin, F., & Yue, Y. (2019) · Signal Processing · doi:10.1016/j.sigpro.2019.107324 · Cited by 21
Robust Structural Low-Rank Tracking
Javed, S., Mahmood, A., Dias, J. M. M., & Werghi, N. (2020) · IEEE Transactions on Image Processing · doi:10.1109/tip.2020.2972102 · Cited by 23
“learn a minimum output sum of squared error filter over the luminance channel for VOT [55].”
BackgroundPart-Based Object Tracking Using Multiple Adaptive Correlation Filters
Barcellos, P., & Scharcanski, J. (2021) · IEEE Transactions on Instrumentation and Measurement · doi:10.1109/tim.2021.3088455 · Cited by 21
Improved Kernel Correlation Filter Based Moving Target Tracking for Robot Grasping
Peng, F., Xu, Q., Li, Y., Zheng, M., & Su, H. (2022) · IEEE Transactions on Instrumentation and Measurement · doi:10.1109/tim.2022.3195258 · Cited by 24
“Classical correlation filtering algorithms include kernelized correlation filter (KCF) [14], MOOSE [15], CSK [16], and CN [17].”
MethodHyperspectral Video Tracker Based on Spectral Deviation Reduction and a Double Siamese Network
Zhang, Z., Hu, B., Wang, M., Arun, P. V., Zhao, D., Zhu, X., Hu, J., Li, H., Zhou, H., & Qian, K. (2023) · Remote Sensing · doi:10.3390/rs15061579 · Cited by 21
“The simple linear update [2,51] is the generally adopted update method.”
MethodSiamRank: A siamese based visual tracking network with ranking strategy
Meng, F., Gong, X., & Zhang, Y. (2023) · Pattern Recognition · doi:10.1016/j.patcog.2023.109630 · Cited by 21
Practical Tracking Method based on Best Buddies Similarity
He, H., Chen, Z., Liu, H., Liu, X., Guo, Y. G., & Li, J. (2023) · Cyborg and Bionic Systems · doi:10.34133/cbsystems.0050 · Cited by 21
“The target tracking method often depends on the definition of a similarity (distance) measurement, such as the mean square error in minimum output sum square error (MOSSE) [7], Bhattacharyya similarity in scale-adaptive meanshift (ASMS) [22], etc. BBS is a nonparametric robust similarity measurement method that counts…”
MethodInfluentialFast and Robust Object Tracking with Adaptive Detection
Bharati, S. P., Nandi, S., Wu, Y., Sui, Y., & Wang, G. (2016) · IEEE International Conference on Tools with Artificial Intelligence (ICTAI) · doi:10.1109/ictai.2016.0112 · Cited by 24
“[28], efficiently trained the correlation filter on gray-scale images, making it computationally efficient and leading to real-time applications.”
BackgroundOnline RGB-D tracking via detection-learning-segmentation
An, N., Zhao, X., & Hou, Z. (2016) · Proceedings - International Conference on Pattern Recognition/Proceedings/International Conference on Pattern Recognition · doi:10.1109/icpr.2016.7899805 · Cited by 21
“The correlation filter based trackers [18], [19], [20], [21], [22] are proposed rencetly to settle the sampling ambiguity problem in binary classifiers.”
BackgroundA hybrid tracking framework based on kernel correlation filtering and particle filtering
Zhao, Z., Feng, P., Guo, J., Yuan, C., Wang, T., Liu, F., Zhao, Z., Cui, Z., & Wu, B. (2018) · Neurocomputing · doi:10.1016/j.neucom.2018.02.043 · Cited by 22
CNN tracking based on data augmentation
Wang, Y., Wei, X., Tang, X., Shen, H., & Ding, L. (2020) · Knowledge-Based Systems · doi:10.1016/j.knosys.2020.105594 · Cited by 23
Visual Tracking With Multiview Trajectory Prediction
Wu, M., Ling, H., Bi, N., Gao, S., Hu, Q., Sheng, H., & Yu, J. (2020) · IEEE Transactions on Image Processing · doi:10.1109/tip.2020.3014952 · Cited by 24
“Based on Discriminative Correlation Filters (DCF), MOSSE is proposed in [10] to efficiently train the correlation filter by minimizing the sum of the squared error between ground-truth and output in the Fourier domain.”
MethodTracking fiducial markers with discriminative correlation filters
Romero-Ramírez, F. J., Muñoz‐Salinas, R., & Medina-Carnicer, R. (2020) · Image and Vision Computing · doi:10.1016/j.imavis.2020.104094 · Cited by 23
Super-Resolution Capacitive Touchscreens
Mayer, S., Xu, X., & Harrison, C. (2021) · CHI Conference on Human Factors in Computing Systems · doi:10.1145/3411764.3445703 · Cited by 24
Multiple Object Tracking with GRU Association and Kalman Prediction
Lit, Z., Cai, S., Wang, X., Shao, H., Niu, L., & Xue, N. (2021) · International Joint Conference on Neural Networks (IJCNN) · doi:10.1109/ijcnn52387.2021.9533828 · Cited by 24
“In order to achieve a better performance in the endto-end learning philosophy, recent works [36], [38] have investigated embedding Discriminative Correlation Filters (DCF) [7], [13] as a computational block in a deep learning network.”
BackgroundA Video Analytics System for Person Detection Combined with Edge Computing
Maltezos, E., Lioupis, P., Dadoukis, A., Karagiannidis, L., Ouzounoglou, E., Krommyda, M., & Amditis, A. (2022) · Computation · doi:10.3390/computation10030035 · Cited by 22
Recent advances in correlation filter theory and application
Kumar, B. V. K. V., Fernandez, J. A., Rodríguez, A., & Boddeti, V. N. (2014) · Proceedings of SPIE, the International Society for Optical Engineering/Proceedings of SPIE · doi:10.1117/12.2051719 · Cited by 21
“4 Minimum output sum of squared error (MOSSE) lter The MOSSE lter was introduced in 2010.(2) This lter is designed to minimize the MSE between the desired correlation plane and the actual correlation plane.”
MethodPatch-based Scale Calculation for Real-time Visual Tracking
Xu, Y., Wang, J., Li, H., Li, Y., Miao, Z., & Zhang, Y. (2015) · IEEE Signal Processing Letters · doi:10.1109/lsp.2015.2479360 · Cited by 21
“We also compare our approachwith 11 state-of-the-art trackers and the results show that our approach outperforms the other methods while running at more than 60 frames per second (FPS).”
MethodNon-negative Subspace Representation Learning Scheme for Correlation Filter Based Tracking
Xu, T., Wu, X., & Kittler, J. (2018) · Proceedings - International Conference on Pattern Recognition/Proceedings/International Conference on Pattern Recognition · doi:10.1109/icpr.2018.8546146 · Cited by 24
“08],β = [4, 6, 8, 10]} for STC with gray intensity values as features.”
BackgroundMethodMotion-Aware Correlation Filters for Online Visual Tracking
Zhang, Y., Yang, Y., Zhou, W., Shi, L., & Li, D. (2018) · Sensors · doi:10.3390/s18113937 · Cited by 20
“In addition, it multiplies a cosine window with the same size of padding region to emphasize on the target [1,14,16,17,26,31].”
BackgroundMethodInfluentialRemote Sensing Video Tracking: Current Status, Challenges, and Future
Liu, F., Wang, J., Jiao, L., Zhang, J., Wang, H., Li, S., Li, L., Chen, P., Liu, X., Ma, W., et al. (2025) · IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing · doi:10.1109/jstars.2025.3573572 · Cited by 23
Spatial windowing for correlation filter based visual tracking
Gündoğdu, E., & Alatan, A. A. (2016) · Proceedings - International Conference on Image Processing · doi:10.1109/icip.2016.7532645 · Cited by 21
“2Although, it should ideally be an impulse function, it is designed as a Gaussian with a small standard deviation to integrate a robustness [3, 5].”
BackgroundMethodInfluentialAdaptive Pulsatile Plane for Robust Noncontact Heart Rate Monitoring
Tran, Q., Su, S., Sun, W., & Tran, M. (2019) · IEEE Transactions on Systems Man and Cybernetics Systems · doi:10.1109/tsmc.2019.2957159 · Cited by 20
Self-supervised Object Tracking with Cycle-consistent Siamese Networks
Yuan, W., Wang, M. Y., & Chen, Q. (2020) · IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS) · doi:10.1109/iros45743.2020.9341621 · Cited by 30
“Recent deep network based methods [6–10] for visual object tracking have dominated most benchmarks and demonstrated advantages over traditional tracking methods [11–16] in both accuracy and speed.”
MethodReal-time implementation of fast discriminative scale space tracking algorithm
Walid, W., Awais, M., Ahmed, A., Masera, G., & Martina, M. (2021) · Journal of Real-Time Image Processing · doi:10.1007/s11554-021-01119-6 · Cited by 21
“Studies have shown that discriminative approaches show better performance and require less computation [6–8].”
BackgroundA Discriminative Single-Shot Segmentation Network for Visual Object Tracking
Lukežič, A., Matas, J., & Kristan, M. (2021) · IEEE Transactions on Pattern Analysis and Machine Intelligence · doi:10.1109/tpami.2021.3137933 · Cited by 20
“erty has been studied in depth in discriminative template trackers called discriminative correlation filters (DCF) [26].”
BackgroundMultibranch Adaptive Fusion Network for RGBT Tracking
Li, Y., Hui-cheng, L., Wang, L., & Jia, Z. (2022) · IEEE Sensors Journal · doi:10.1109/jsen.2022.3154657 · Cited by 22
Learning Spatio-Temporal Attention Based Siamese Network for Tracking UAVs in the Wild
Chen, J., Huang, B., Li, J., Wang, Y., Ren, M., & Xu, T. (2022) · Remote Sensing · doi:10.3390/rs14081797 · Cited by 21
“MOSSE [17] tracker is the first to validate the feasibility of the correlation filter in tracking.”
MethodAlpine Skiing Tracking Method Based on Deep Learning and Correlation Filter
Qi, J., Li, D., Zhang, C., & Wang, Y. (2022) · IEEE Access · doi:10.1109/access.2022.3166949 · Cited by 20
“In 2010, the minimum output sum of squared error filter (MOSSE) [16] was the first method using the idea of correlation filtering in visual tracking.”
MethodSpike-Event Object Detection for Neuromorphic Vision
Wang, Y.-K., Wang, S.-E., & Wu, P. (2023) · IEEE Access · doi:10.1109/access.2023.3236800 · Cited by 20
“Several advanced tracking algorithms, such as Discriminative Correlation Filter with Channel and Spatial Reliability (CSR-DCF) [55], Kernelized Correlation Filter (KCF) [60], and Minimum Output Sum of Squared Error [61], can be employed in our framework.”
MethodAn Active Multi-Object Ultrafast Tracking System with CNN-Based Hybrid Object Detection
Li, Q., Hu, S., Shimasaki, K., & Ishii, I. (2023) · Sensors · doi:10.3390/s23084150 · Cited by 23
“Early classical tracking methods, such as Meanshift [59], particle filtering [60], KCF [61], and MOSSE [62], mainly focused on single-object tracking.”
MethodInfluentialComputer Vision: Advances in Image and Video Analysis
Dhabliya, D., Ugli, I. S. M., Murali, M. J., Abbas, A. H. R., & Gulbahor, U. (2023) · E3S Web of Conferences · doi:10.1051/e3sconf/202339904045 · Cited by 20
“The research also looks at new developments in object tracking, such domain adaption and online learning.[10][11][12] [34] Review of Deep Learning for Image Super-Resolution In order to improve the resolution and details of low-resolution photos, this review article investigates the use of deep learning in image super…”
BackgroundSearching Region-Free and Template-Free Siamese Network for Tracking Drones in TIR Videos
Huang, B., Dou, Z., Chen, J., Li, J., Shen, N., Wang, Y., & Xu, T. (2023) · IEEE Transactions on Geoscience and Remote Sensing · doi:10.1109/tgrs.2023.3341331 · Cited by 30
“Bolme et al. [26] first introduce correlation filters into the tracking field by presenting a minimum output sum of squared error (MOSSE) filter that enables tracking with an extremely high speed of 669 frames/s. Inspired by MOSSE, a vast number of followers have improved the performance of…”
BackgroundTemporal Correlation Vision Transformer for Video Person Re-Identification
Wu, P., Wang, L., Zhou, S., Hua, G., & Sun, C. (2024) · Proceedings of the AAAI Conference on Artificial Intelligence · doi:10.1609/aaai.v38i6.28424 · Cited by 31
“Bolme et al. (Bolme et al. 2010) learn a minimum output sum of the squared error filter and update it with average moving.”
BackgroundDepth Masked Discriminative Correlation Filter
Kart, U., Kämäräinen, J., Matas, J., Fan, L., & Cricri, F. (2018) · Proceedings - International Conference on Pattern Recognition/Proceedings/International Conference on Pattern Recognition · doi:10.1109/icpr.2018.8546179 · Cited by 25
Augmenting cascaded correlation filters with spatial–temporal saliency for visual tracking
Zhao, D., Xiao, L., Fu, H., Wu, T., Xu, X., & Dai, B. (2018) · Information Sciences · doi:10.1016/j.ins.2018.08.053 · Cited by 20
Multi-level context-adaptive correlation tracking
Liu, P., Liu, C., Zhao, W., & Tang, X. (2018) · Pattern Recognition · doi:10.1016/j.patcog.2018.10.013 · Cited by 20
“Correlation filters were introduced to visual tracking by MOSSE [11], and a remarkably high speed was obtained using the fast Fourier transform (FFT).”
BackgroundMethodInfluentialCorrelation Filter Based Moving Object Tracking With Scale Adaptation and Online Re-Detection
Islam, M. M., Hu, G., Liu, Q., Wang, D., & Lyu, C. (2018) · IEEE Access · doi:10.1109/access.2018.2883650 · Cited by 20
“Contrary with the KCF tracker, the MOSSE method [40] introduced a robust model updating technique to update the present model by considering all the past frames.”
MethodInfluentialCorrelation Filter Tracking via Distractor-Aware Learning and Multi-Anchor Detection
Chen, G., Pan, G., Zhou, Y., Kang, W., Hou, J., & Deng, F. (2019) · IEEE Transactions on Circuits and Systems for Video Technology · doi:10.1109/tcsvt.2019.2961999 · Cited by 20
DroTrack: High-speed Drone-based Object Tracking Under Uncertainty
Ali, A., Salim, F. D., & Kim, D. Y. (2020) · RMIT Research Repository (RMIT University Library) · Cited by 20
Learning a Dynamic Feature Fusion Tracker for Object Tracking
Li, Z., Nai, K., Li, G., & Jiang, S. (2020) · IEEE Transactions on Intelligent Transportation Systems · doi:10.1109/tits.2020.3027521 · Cited by 21
Correlation Discrepancy Insight Network for Video Re-identification
Ruan, W., Liang, C., Yu, Y., Wang, Z., Liu, W., Chen, J., & Ma, J. (2020) · ACM Transactions on Multimedia Computing Communications and Applications · doi:10.1145/3402666 · Cited by 20
“Although this process is operated frame by frame, it requires little time due to the high efficiency of CF, which can handle hundreds of frames per second, as demonstrated in object tracking [4, 13].”
BackgroundLearning spatio-temporal correlation filter for visual tracking
Yan, Y., Guo, X., Tang, J., Li, C., & Wang, X. (2021) · Neurocomputing · doi:10.1016/j.neucom.2021.01.057 · Cited by 20
Learning Channel-Aware Correlation Filters for Robust Object Tracking
Nai, K., Li, Z., & Wang, H. (2022) · IEEE Transactions on Circuits and Systems for Video Technology · doi:10.1109/tcsvt.2022.3186276 · Cited by 21
“Correlation Filters (CFs) based trackers [1]–[19] have achieved encouraging tracking accuracy and speeds in object tracking field.”
BackgroundMethodInfluentialFeature selection accelerated convolutional neural networks for visual tracking
Cui, Z., & Lü, N. (2021) · Applied Intelligence · doi:10.1007/s10489-021-02234-4 · Cited by 20
“MOSSE (Minimum Output Sum of Squared Error) [13] was the first algorithm to use CF to track objects.”
MethodAn, Z., Wang, X., Li, B., & Fu, J. (2023) · IEEE Transactions on Instrumentation and Measurement · doi:10.1109/tim.2023.3265106 · Cited by 19
“Popular failure detection methods [17], [26], [29] focus on the response map formed by the target feature and determine the target state [i.e., illumination variation (IV) and low resolution (LR)] by calculating the peak intensity or fluctuation of the response map.”
BackgroundMethodInfluentialUAV Tracking Based on Correlation Filters With Dynamic Aberrance-Repressed Temporal Regularizations
Zhang, H., Li, Y., Yang, Y. F., Feng, Y., Li, Y., Deng, C., & Yuan, D. (2023) · IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing · doi:10.1109/jstars.2023.3306273 · Cited by 19
“[22] first proposed the correlation filter (CF) method for visual object tracking, which derives the minimum output sum of squared error between the ideal given response and the correlation response.”
MethodDeep features fusion for KCF-based moving object tracking
Maharani, D. A., Machbub, C., Yulianti, L., & Rusmin, P. H. (2023) · Journal Of Big Data · doi:10.1186/s40537-023-00813-5 · Cited by 19
“[5] introduced the minimum output sum of squared error (MOSSE), the first object-tracking method to use correlation filtering.”
BackgroundMethodSFA-guided mosaic transformer for tracking small objects in snapshot spectral imaging
Chen, L., Zhao, Y., & Kong, S. G. (2023) · ISPRS Journal of Photogrammetry and Remote Sensing · doi:10.1016/j.isprsjprs.2023.09.015 · Cited by 19
Knowledge Synergy Learning for Multi-Modal Tracking
He, Y., Ma, Z., Wei, X., & Gong, Y. (2024) · IEEE Transactions on Circuits and Systems for Video Technology · doi:10.1109/tcsvt.2024.3352573 · Cited by 21
“The Correlation Filter (CF) based trackers [20], [21], [22], [23] have attracted extensive attention in the past decades.”
BackgroundMethodDivert More Attention to Vision-Language Object Tracking
Guo, M., Zhang, Z., Jing, L., Ling, H., & Fan, H. (2024) · IEEE Transactions on Pattern Analysis and Machine Intelligence · doi:10.1109/tpami.2024.3409078 · Cited by 28
“The seminal work MOSSE [29] first introduces the correlation operation in the frequency domain into visual tracking, motivating a series of CF-based works [30], [31], [32], [33].”
BackgroundTracking With Static and Dynamic Structured Correlation Filters
Wang, S., Wang, D., & Lu, H. (2017) · IEEE Transactions on Circuits and Systems for Video Technology · doi:10.1109/tcsvt.2017.2714429 · Cited by 19
Kim, Y.-G., & Huynh, X.-P. (2017) · IEEE International Conference on Computer Vision Workshops · doi:10.1109/iccvw.2017.362 · Cited by 20
“MOSSE (Minimum Output Sum of Squared Error) tracker is known to be robust against variations of illumination, scale and pose while operating at high speed [1].”
BackgroundToward Occlusion Handling in Visual Tracking via Probabilistic Finite State Machines
Liu, C., Huynh, D. Q., & Reynolds, M. A. (2018) · IEEE Transactions on Cybernetics · doi:10.1109/tcyb.2018.2884007 · Cited by 19
“[17] is the seminal paper in correlation filter-based tracking.”
BackgroundMethodTracking-by-Fusion via Gaussian Process Regression Extended to Transfer Learning
Gao, J. G., Wang, Q., Xing, J., Ling, H., Hu, W., & Maybank, S. J. (2018) · IEEE Transactions on Pattern Analysis and Machine Intelligence · doi:10.1109/tpami.2018.2889070 · Cited by 21
“[54] develop a MinimumOutput Sum of Squared Error (MOSSE) filter, and use a single feature channel, typically an intensity feature channel, for tracking.”
BackgroundMethodAerial Infrared Target Tracking in Complex Background Based on Combined Tracking and Detecting
Hu, Y., Xiao, M., Zhang, K., & Wang, X. (2019) · Mathematical Problems in Engineering · doi:10.1155/2019/2419579 · Cited by 22
“In [25], the correlation filters could deal with 669 figures per second, which raises the speed of tracking algorithm from the real-time level to the highspeed level.”
BackgroundZolfaghari, M., Ghanei-Yakhdan, H., & Yazdi, M. (2019) · The Visual Computer · doi:10.1007/s00371-019-01652-3 · Cited by 20
“Bolme et al. [7] proposed a new type of correlation filter, namely the minimum output sum of square error (MOSSE), which performs well under partial occlusion, illumination and rotation changes.”
BackgroundOne-Shot Scale and Angle Estimation for Fast Visual Object Tracking
Lee, D. (2019) · IEEE Access · doi:10.1109/access.2019.2913390 · Cited by 19
“filter-based trackers train a regressor to discriminate the template image patch, where the target object is centered, from the surrounding patches by solving a large ridge regression problem [3], [4].”
BackgroundRemote Sensing Target Tracking in UAV Aerial Video Based on Saliency Enhanced MDnet
Bi, F., Lei, M., Wang, Y., & Huang, D. F. (2019) · IEEE Access · doi:10.1109/access.2019.2921315 · Cited by 20
FAST: Fast and Accurate Scale Estimation for Tracking
Ma, H., Lin, Z., & Acton, S. T. (2019) · IEEE Signal Processing Letters · doi:10.1109/lsp.2019.2963147 · Cited by 19
“Recently, the discriminative correlation filter (DCF) has been successfully applied to VOT yielding both high accuracy and high efficiency [7]–[9], [34]–[36].”
MethodA Scale-Adaptive Matching Algorithm for Underwater Acoustic and Optical Images
Liu, J., Li, B., Guan, W., Gong, S., Liu, J., & Cui, J. (2020) · Sensors · doi:10.3390/s20154226 · Cited by 21
“Bolme introduced the correlation filter into target tracking for the first time and proposed a method using calculation of the correlation [18].”
MethodMulti-Task Convolution Operators With Object Detection for Visual Tracking
Zheng, Y., Liu, X., Xiao, B., Cheng, X., Wu, Y., & Chen, S. (2021) · IEEE Transactions on Circuits and Systems for Video Technology · doi:10.1109/tcsvt.2021.3071128 · Cited by 20
“Visual tracking approaches can be roughly divided into two categories: correlation filter (CF) methods [1]–[8] and deep learning (DL) methods [9]–[14].”
BackgroundMethodResultInfluentialLearning Saliency-Aware Correlation Filters for Visual Tracking
Wang, Y., Wang, F., Wang, C., Sun, F., & He, J. (2021) · The Computer Journal · doi:10.1093/comjnl/bxab026 · Cited by 20
Towards a Comprehensive and Robust Micromanipulation System with Force-Sensing and VR Capabilities
Adam, G., Chidambaram, S., Reddy, S. S., Ramani, K., & Cappelleri, D. J. (2021) · Micromachines · doi:10.3390/mi12070784 · Cited by 19
“MOSSE was the fastest method, but was not reliable enough to provide robust tracking.”
MethodLearning Future-Aware Correlation Filters for Efficient UAV Tracking
Zhang, F., Ma, S., Yu, L., Zhang, Y., Qiu, Z., & Li, Z. (2021) · Remote Sensing · doi:10.3390/rs13204111 · Cited by 20
Andersson, O., Doherty, P., Lager, M., Lindh, J.-O., Persson, L., Topp, E. A., Tordenlid, J., & Wahlberg, B. (2021) · Autonomous Intelligent Systems · doi:10.1007/s43684-021-00009-9 · Cited by 19
“We evaluated several tracking approaches such as Boosting [71], MIL [72], KCF [73], TLD [74],MedianFlow [75],Mosse [76], and CSRT (OpenCV implementation of CSR-DCF [77]) • Align information from detection and tracking steps with that given in the sea chart and propose an annotation for the found and tracked object • V…”
MethodLearning Spatio-Appearance Memory Network for High-Performance Visual Tracking
Xie, F., Yang, W., Zhang, K., Liu, B., Wang, G., & Zuo, W. (2021) · IEEE International Conference on Computer Vision Workshops · doi:10.1109/iccvw54120.2021.00302 · Cited by 23
“Later, D3S [27] introduces a segmentation branch following VideoMatch [13] and further combines online DCF [2] to fuse target classification and pixel-wise segmentation during inference.”
MethodInfluentialSiamOAN: Siamese object-aware network for real-time target tracking
Wei, B., Chen, H., Ding, Q., & Luo, H. (2021) · Neurocomputing · doi:10.1016/j.neucom.2021.10.112 · Cited by 19
DASFTOT: Dual attention spatiotemporal fused transformer for object tracking
Wu, R., Wen, X., Yuan, L., & Xu, H. (2022) · Knowledge-Based Systems · doi:10.1016/j.knosys.2022.109897 · Cited by 19
UAV Chasing Based on YOLOv3 and Object Tracker for Counter UAV Systems
Kim, K., Kim, J., Lee, H.-G., Choi, J., Fan, J., & Joung, J. (2023) · IEEE Access · doi:10.1109/access.2023.3264603 · Cited by 19
“The drawback of the MOSSE tracker is that it does not guarantee accurate tracking performance when the object’s appearance changes [30].”
BackgroundDCFNet: Discriminant Correlation Filters Network for Visual Tracking
Hu, W., Wang, Q., Gao, J. G., Li, B., & Maybank, S. J. (2024) · Journal of Computer Science and Technology · doi:10.1007/s11390-023-3788-3 · Cited by 20
“MOSSE [6] first introduces DCF to visual tracking, which runs at high speed of more than 600 FPS simply using the single-channel gray features.”
MethodUnmanned Aerial Vehicle Object Tracking by Correlation Filter with Adaptive Appearance Model
Xue, X., Li, Y., & Shen, Q. (2018) · Sensors · doi:10.3390/s18092751 · Cited by 19
“[18] have proposed the minimum output sum of squared errors (MOSSE) filter, which works by trying to find the maximum cross correlation response between the model and candidate patch.”
BackgroundCorrelation-Based Tracker-Level Fusion for Robust Visual Tracking
Rapuru, M. K., Kakanuru, S., Venugopal, P. M., Mishra, D., & Gorthi, R. K. (2017) · IEEE Transactions on Image Processing · doi:10.1109/tip.2017.2699791 · Cited by 18
“Recently, correlation filters [4], [15], [16] have proved to be competitive”
BackgroundRotation invariant person tracker using top view
Ullah, K., Ahmed, I., Ahmad, M., Rahman, A. U., Nawaz, M., & Adnan, A. (2019) · Journal of Ambient Intelligence and Humanized Computing · doi:10.1007/s12652-019-01526-5 · Cited by 21
“MOSSE fails to track the person when there is a change in direction or pose.”
Learning Temporal-Correlated and Channel- Decorrelated Siamese Networks for Visual Tracking
Xi, M., Zhou, W., Wang, N., & Li, H. (2021) · IEEE Transactions on Multimedia · doi:10.1109/tmm.2021.3087340 · Cited by 18
Long-Term Person Tracking for Unmanned Aerial Vehicle Based on Human-Machine Collaboration
Zhou, T., & Liu, Y. (2021) · IEEE Access · doi:10.1109/access.2021.3132077 · Cited by 19
“Most studies of visual tracking are based on the assumption that the tracking target changes smoothly and does not disappear for a long time [1-4].”
BackgroundIntravascular Tracking of Micro-Agents Using Medical Ultrasound: Towards Clinical Applications
Šuligoj, F., Heunis, C. M., Mohanty, S., & Misra, S. (2022) · IEEE Transactions on Biomedical Engineering · doi:10.1109/tbme.2022.3176746 · Cited by 18
A model adaptive updating kernel correlation filter tracker with deep CNN features
Feng, Z., & Wang, P. (2023) · Engineering Applications of Artificial Intelligence · doi:10.1016/j.engappai.2023.106250 · Cited by 19
Plattner, M., & Ostermayer, G. (2023) · Vehicular Communications · doi:10.1016/j.vehcom.2023.100651 · Cited by 18
Zhu, Q., Huang, X., & Guan, Q. (2024) · International Journal of Applied Earth Observation and Geoinformation · doi:10.1016/j.jag.2024.103723 · Cited by 23
Leveraging the Power of Data Augmentation for Transformer-based Tracking
Zhao, J. W., Edstedt, J., Felsberg, M., Wang, D., & Lu, H. (2024) · IEEE Winter Conference on Applications of Computer Vision · doi:10.1109/wacv57701.2024.00634 · Cited by 23
“traditional approaches [5, 18, 20] using hand-crafted features, CNNbased methods [1, 2, 23, 47], and transformer-based methods [7, 9, 27, 41, 43].”
MethodAn anti-occlusion vision-based method for structural motion estimation
Hou, J., Zhang, Y., Lu, X., Cai, E., Wei, K., Luo, M., Guo, J. H., Ma, Z., Hoon Sohn, A., & Guo, T. (2024) · Mechanical Systems and Signal Processing · doi:10.1016/j.ymssp.2024.112003 · Cited by 18
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.