Multiple Cues-Based Robust Visual Object Tracking Method
Open accessThe DOI matches its Crossref record; 37 of the 37 references it cites were found in a registry or an index; 22 of the 52 sources its claims cite were checked for support.
Abstract
Visual object tracking is still considered a challenging task in computer vision research society. The object of interest undergoes significant appearance changes because of illumination variation, deformation, motion blur, background clutter, and occlusion. Kernelized correlation filter- (KCF) based tracking schemes have shown good performance in recent years. The accuracy and robustness of these trackers can be further enhanced by incorporating multiple cues from the response map. Response map computation is the complementary step in KCF-based tracking schemes, and it contains a bundle of information. The majority of the tracking methods based on KCF estimate the target location by fetching a single cue-like peak correlation value from the response map. This paper proposes to mine the response map in-depth to fetch multiple cues about the target model. Furthermore, a new criterion based on the hybridization of multiple cues i.e., average peak correlation energy (APCE) and confidence of squared response map (CSRM), is presented to enhance the tracking efficiency.
Overview
CoChat Summary
TL;DR
This paper proposes a KCF-based visual object tracker that mines multiple cues from the correlation response map—APCE, CSRM, peak correlation, and peak-to-sidelobe ratio—to handle occlusion, scale variation, drift, and other challenges. On 23 selected videos from OTB-50, OTB-100, and TC-128, it reports the highest mean distance precision (0.889), overlap success rate (0.824), and lowest mean center location error (16.06) among compared trackers.
Key findings
- Mean distance precision at 20 pixels: proposed 0.889, versus MKCF 0.855, MACF 0.804, STRCF 0.784, and LCT 0.761.
- Mean overlap success rate: proposed 0.824, versus MACF 0.822, STRCF 0.799, LCT 0.717, and MKCF 0.645.
- Mean center location error: proposed 16.06, versus STRCF 44.89, MACF 53.94, LCT 70.797, and MKCF 76.782 in Table 4; the text reports MKCF as 79.782.
- Qualitative results: in the ball_ce3 sequence, after the ball returns to view at frame 256, the proposed tracker is the only one shown to track it correctly; it also handles car1, microbike_ce, railwaystation_ce, and suitcase_ce sequences.
Methods
- Baseline KCF tracker extended with multiple response-map cues: APCE is used to select the true object position, a multi-resolution scale search selects scale when CSRM and APCE exceed a threshold, and CSRM adjusts the learning rate; occlusion detection uses hybridization of APCE and CSRM, with threshold Th set to 0.6.
- Evaluation uses 23 selected videos from OTB-50, OTB-100, and TC-128, covering 11 challenge attributes, and compares against LCT, MACF, MKCF, and STRCF using distance precision at 20 pixels, center location error, and overlap success rate at 0.5.
- Implementation is in MATLAB 2019 on an Intel Core i7 7th generation 2.80 GHz processor with 16 GB RAM and 64-bit Windows 10.
Limitations
- The paper does not state a limitations section; the Data Availability Statement says “Not applicable.”
- Evident from the text: evaluation is limited to 23 selected sequences rather than the full OTB-50, OTB-100, and TC-128 benchmarks, and no runtime, FPS, or computational-cost measurements are reported despite the paper’s emphasis on efficiency.
- Evident from the text: there is an inconsistency in the reported mean center location error for MKCF (79.782 in the text versus 76.782 in Table 4), and qualitative analysis is limited to selected sequences and frames.
Model-written summary. Check it against the paper before relying on it.
Integrity signals
Registry
Cites
Statistics
Verification
Claims the paper makes
14 Valid · 8 Not valid · 19 Not checked · 26 Own results and interpretations
Every source the paper cites for this claim was checked and supports it.
In the paper
“In contrast, MACF [23] is second-best with a small difference of 0.002”
What the sources say
- Supportedcites [23] Zhang, Y., Yang, Y., Zhou, W., Shi, L., and Li, D. (2018). Motion-Aware Correlation Filters for Online Visual Tracking. Sensors, 18. source
The citation identifies MACF, and the source paper is the work introducing that tracker. The comparison with the citing paper’s proposed tracker is the citing authors’ own result, not a finding attributed to this source. (LLM, conf=0.90)
plain- Asserted: passes — The authors put it forward as true.
- Checkable: passes — It can be checked against evidence.
- Load-bearing: unclear — It is unclear whether the paper’s conclusions depend on it.
- Based: passes — The paper gives a reason why it holds.
- Supported
The cited sources that could be checked couldn’t confirm or dispute this claim.
In the paper
“whereas the third-best, i.e., MACF [23], achieved the mean error of 53.94 pixels.”
What the sources say
- Could not verifycites [23] Zhang, Y., Yang, Y., Zhou, W., Shi, L., and Li, D. (2018). Motion-Aware Correlation Filters for Online Visual Tracking. Sensors, 18. sourceExperimental — accuracy not yet measured
The paper’s Conclusions report MACF precision and success results, but the passages found do not give a mean center-location error of 53.94 pixels or rank MACF third by that measure. Searches for “53.94” and mean center-location error did not locate the claimed result in our copy. (LLM, conf=0.73)
plain- Asserted: passes — The authors put it forward as true.
- Checkable: passes — It can be checked against evidence.
- Load-bearing: unclear — It is unclear whether the paper’s conclusions depend on it.
- Based: passes — The paper gives a reason why it holds.
- Could not verify
The cited sources that could be checked couldn’t confirm or dispute this claim.
In the paper
“whereas the third best is MACF [23], having a mean value of 0.804.”
What the sources say
- Could not verifycites [23] Zhang, Y., Yang, Y., Zhou, W., Shi, L., and Li, D. (2018). Motion-Aware Correlation Filters for Online Visual Tracking. Sensors, 18. sourceExperimental — accuracy not yet measured
The paper’s precision plots (section 5, page not specified) evaluate MACF, but the retrieved results do not report a mean distance precision of 0.804 or a third-place ranking. I searched for the value and for distance-precision rankings; the paper’s results concern its own tracker comparisons, so they do not establish the cited sentence’s ranking. (LLM, conf=0.74)
plain- Asserted: passes — The authors put it forward as true.
- Checkable: passes — It can be checked against evidence.
- Load-bearing: unclear — It is unclear whether the paper’s conclusions depend on it.
- Based: passes — The paper gives a reason why it holds.
- Could not verify
The cited sources support only part of this claim — read each source’s verdict below.
In the paper
“while the remaining three trackers, i.e., STRCF [28], LCT [21], and MKCF [5], have a mean overlap success rate of 0.799, 0.717, and 0.645, respectively.”
What the sources say
- Partly supportedcites [21] Ma, 2018, Adaptive Correlation Filters with Long-Term and Short-Term Memory for Object Tracking, Int. J. Comput. Vis. sourceExperimental — accuracy not yet measured
The Introduction (p. 2) describes the authors’ tracker, supporting the citation’s identification of LCT. The overlap-success results in Overall Performance (pp. 12–13) do not report a mean rate of 0.717 for it, so the numerical part of the claim is not supported by our copy. (LLM, conf=0.83)
plain- Asserted: passes — The authors put it forward as true.
- Checkable: passes — It can be checked against evidence.
- Load-bearing: unclear — It is unclear whether the paper’s conclusions depend on it.
- Based: passes — The paper gives a reason why it holds.
- Partly supported
The cited sources that could be checked couldn’t confirm or dispute this claim.
In the paper
“LCT [21] and MKCF [5] showed similar performance in terms of center location error by achieving 70.797 and 79.782 values, respectively.”
What the sources say
- Could not verifycites [21] Ma, 2018, Adaptive Correlation Filters with Long-Term and Short-Term Memory for Object Tracking, Int. J. Comput. Vis. sourceExperimental — accuracy not yet measured
Section 6.8 (p. 18) says the paper compares trackers’ center location errors, and the Figure 19 caption (p. 19) describes those results. Neither passage gives a mean error of 70.797 pixels for LCT. Searches for the exact value, LCT, and mean center location error found no passage verifying the claimed number. (LLM, conf=0.74)
plain- Asserted: passes — The authors put it forward as true.
- Checkable: passes — It can be checked against evidence.
- Load-bearing: unclear — It is unclear whether the paper’s conclusions depend on it.
- Based: passes — The paper gives a reason why it holds.
- Could not verify
29 more claims · 3 with their cited sources checked
Create a free account to see all 67 claims and to verify the 1 reference still open.
- Probably not a substantial claim. 22
- It may state two separate claims as one. 13
- The quote only partly states this claim. 3
- Borderline: one of the tests for a substantial claim is unclear. 1
Self-check
Percentages, sums and reported differences re-computed from the paper’s own numbers.
Each contradiction between two passages listed here was confirmed by an independent check.
Rated minor: it would not touch a result — a value no result rests on, a p-value that stays on the same side of the significance level, or a conflict that is clerical, resolved elsewhere in the paper, or outside its own findings.
In the paper
§ 5.2. Qualitative Analysis
“For the qualitative analysis, the results of five trackers, i.e., our proposed, MKCF [5], MACF [23], STRCF [28], and LCT [21], over five video sequences are presented in Figure 5.”
§ 5.2. Qualitative Analysis
“Figure 5. Qualitative analysis over six selected videos from OTB-50, OTB-100, and Colour-128 datasets.”
Rated minor: it would not touch a result — a value no result rests on, a p-value that stays on the same side of the significance level, or a conflict that is clerical, resolved elsewhere in the paper, or outside its own findings.
In the paper
§ 6. Conclusions
“Funding: This research received no external funding.”
§ References
“Acknowledgments: The authors extend their appreciation to King Saud University for funding this work through the Researcher Support Project number (RSP-2021/133), King Saud University, Riyadh, Saudi Arabia.”
Rated minor: it would not touch a result — a value no result rests on, a p-value that stays on the same side of the significance level, or a conflict that is clerical, resolved elsewhere in the paper, or outside its own findings.
In the paper
§ 5.1. Quantitative Analysis
“LCT [21] and MKCF [5] showed similar performance in terms of center location error by achieving 70.797 and 79.782 values, respectively.”
§ 5.1. Quantitative Analysis
“| Mean | 16.0600 | 70.7970 | 53.9400 | 76.7820 | 44.8900 |”
Rated minor: it would not touch a result — a value no result rests on, a p-value that stays on the same side of the significance level, or a conflict that is clerical, resolved elsewhere in the paper, or outside its own findings.
In the paper
§ 3.3. Adaptive Scale Handling Mechanism 3.3. Adaptive Scale Handling Mechanism
“On the other side, when the video contains several challenging factors such as occlusion, motion blur, and scale variation, response map values will start fluctuating, and the APCE measure will decrease.”
§ 3.4. Adaptive Learning Rate
“if the target apparently appears in the detection scope, there will be a sharper peak in the response map, and the value of APCE will be smaller. However, if the target is occluded, the peak in the response map will be smoother, and the relative value of APCE becomes larger [7].”
Citations
Cites
High-Speed Tracking with Kernelized Correlation Filters
Henriques, J. F., Caseiro, R., Martins, P., & Batista, J. (2015) · IEEE Transactions on Pattern Analysis and Machine Intelligence · doi:10.1109/tpami.2014.2345390 · Cited by 5,753
Online Object Tracking: A Benchmark
Wu, Y., Lim, J., & Yang, M.-H. (2013) · 2013 IEEE Conference on Computer Vision and Pattern Recognition · doi:10.1109/cvpr.2013.312 · Cited by 4,138 · cited 8×
“Twenty-three videos were selected from three standard datasets, i.e., OTB-50 [8], OTB-100 [9], and Temple Colour-128 [10].”
InfluentialVisual object tracking using adaptive correlation filters
Bolme, D., Beveridge, J. R., Draper, B. A., & Lui, Y. M. (2010) · 2010 IEEE Computer Society Conference on Computer Vision and Pattern Recognition · doi:10.1109/cvpr.2010.5539960 · Cited by 3,406 · cited 1×
“The initial correlation filter-based tracking proposed in [16] became very famous.”
BackgroundWu, Y., Lim, J., & Yang, M.-H. (2015) · IEEE Transactions on Pattern Analysis and Machine Intelligence · doi:10.1109/tpami.2014.2388226 · Cited by 3,371 · cited 9×
“Twenty-three videos were selected from three standard datasets, i.e., OTB-50 [8], OTB-100 [9], and Temple Colour-128 [10].”
InfluentialIncremental 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 2×
“This problem is solved by squaring the response map and then finding the confidence of the squared response map [7].”
BackgroundMethodExploiting the Circulant Structure of Tracking-by-Detection with Kernels
Henriques, J. F., Caseiro, R., Martins, P., & Batista, J. (2012) · Lecture Notes in Computer Science · doi:10.1007/978-3-642-33765-9_50 · Cited by 2,310
Robust object tracking with online multiple instance learning
Babenko, B., Yang, M.-H., & Belongie, S. (2011) · IEEE Transactions on Pattern Analysis and Machine Intelligence · doi:10.1109/tpami.2010.226 · Cited by 2,007
Struck: Structured output tracking with kernels
Hare, S., Saffari, A., & Torr, P. H. S. (2011) · 2011 International Conference on Computer Vision · doi:10.1109/iccv.2011.6126251 · Cited by 1,831
Discriminative Scale Space Tracking
Danelljan, M., Hager, G., Khan, F. S., & Felsberg, M. (2017) · IEEE Transactions on Pattern Analysis and Machine Intelligence · doi:10.1109/tpami.2016.2609928 · Cited by 1,315 · cited 1×
“In [18], the author proposed adaptive multi-scale correlation filter-based tracking to address the scale variation problem, which exists in the original KCF scheme.”
BackgroundRobust object tracking via sparsity-based collaborative model
Zhong, W., Lu, H., & Yang, M.-H. (2012) · 2012 IEEE Conference on Computer Vision and Pattern Recognition · doi:10.1109/cvpr.2012.6247882 · Cited by 990
Research works printed but not linked: 1
- Mei, X., and Ling, H. (October, January 29). Robust visual tracking using ℓ1minimization. Proceedings of the 2009 IEEE 12th International Conference on Computer Vision, Kyoto, Japan.
Cited by
Deep 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
“These filter-based tracking methods are well-known for their computational efficiency, making them particularly valuable for real-time applications [6, 7].”
BackgroundSpace dynamic target tracking method based on five-frame difference and Deepsort
Huang, C., Zeng, Q., Xiong, F., & Xu, J. (2024) · Scientific Reports · doi:10.1038/s41598-024-56623-z · Cited by 12
An improved spatial–temporal regularization method for visual object tracking
Hayat, M. U., Ali, A., Khan, B., Mehmood, K., Ullah, K., & Amir, M. (2023) · Signal Image and Video Processing · doi:10.1007/s11760-023-02842-2 · Cited by 7
“The hybrid model based on learning criterion is presented for occlusion detection, handling, and scale estimation to bring effectiveness in tracking [28].”
Self-citationWei, H., Zhai, X., & Wu, H. (2024) · IEEE Access · doi:10.1109/access.2024.3365501 · Cited by 5
“Khan et al. [10] proposed a new criterion based on the hybridization of multiple cues i.e., average peak correlation energy (APCE) and confidence of squared response map (CSRM), which is presented to enhance the tracking efficiency.”
BackgroundHSV-Template Matching With MEESPSO Algorithm for Human Tracking in a Crowded Environment
Fauzi, N. I. H., & Musa, Z. B. (2024) · IEEE Access · doi:10.1109/access.2024.3448242 · Cited by 2
“Additionally, the object that moves in any direction may cause deformation to occur [30], [31], [32].”
Background
Metadata
- Authors
- Khan, B., Jalil, A., Ali, A., Alkhaledi, K., Mehmood, K., Cheema, K. M., Murad, M., Tariq, H., & El-Sherbeeny, A. M.
- Affiliations
- International Islamic University, Islamabad (PK)
- Bahria University (PK)
- Kuwait University (KW)
- Khwaja Fareed University of Engineering and Information Technology (PK)
- Gdańsk University of Technology (PL)
- Industrial Engineering Department, College of Engineering, King Saud University, P.O. Box 800, Riyadh 11421, Saudi Arabia
- Corresponding author
- Cheema, Khalid Mehmood
- ORCID
- Venue
- Electronics
- Publisher
- MDPI AG
- Published
- 2022-01-24
- Volume
- 11
- Issue
- 3
- Pages
- 345
- ISSN
- 2079-9292 (ISSN-L)
- OpenAlex
- W4210548271
- Cited by
- 5
- References
- 36
- Citations by year
- 2024: 3 · 2023: 2
- FWCI
- 0.41
- Type
- journal article
- Language
- en
- Open access
- gold
- doi.org
- Licence
- Locations
- Electronics (publishedVersion, cc-by)
- DOAJ (DOAJ: Directory of Open Access Journals) (submittedVersion, cc-by-sa)
- Funders
- King Saud University (RSP-2021/133)
- 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 › Visual Attention and Saliency Detection
- Keywords
- visual object tracking, kernelized correlation filter, response map, multiple cues, occlusion detection, adaptive learning rate, scale estimation, appearance variation
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.
The claim verdicts this report counts are experimental: their accuracy is not yet measured.
A language model drew the claims from the paper on 2026-10-08, under version 3 of the claim definition (model deepseek-v4.1-flash, prompt extract-v3).
A second model labelled each claim on 2026-10-08 (model jev-1.13-20260917, definition revision gold-guideline-v0.2).
Cited sources are checked with a language model’s help (gpt-6-sol).
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.