Vehicle Tracking Using Deep SORT with Low Confidence Track Filtering
The DOI matches its Crossref record; 21 of the 21 references it cites were found in a registry or an index.
Abstract
Multi-object tracking (MOT) becomes an attractive topic due to its wide range of usability in video surveillance and traffic monitoring. Recent improvements on MOT has focused on tracking-by-detection manner. However, as a relatively complicated and integrated computer vision mission, state-of-the-art tracking-by-detection techniques are still suffering from issues such as a large number of false-positive tracks. To reduce the effect of unreliable detections on vehicle tracking, in this paper, we propose to incorporate a low confidence track filtering into the Simple Online and Realtime Tracking with a Deep association metric (Deep SORT) algorithm. We present a self-generated UA-DETRAC vehicle re-identification dataset which can be used to train the convolutional neural network of Deep SORT for data association. We evaluate our proposed tracker on UA-DETRAC test dataset. Experimental results show that the proposed method can improve the original Deep SORT algorithm with a significant margin. Our tracker outperforms the state-of-the-art online trackers and is comparable with batch-mode trackers.
Overview
Integrity signals
Registry
Cites
Statistics
Verification
Claims the paper makes
Claims other papers make about this one
Each is checked against this paper’s own text: does it say what the citing paper says it does?
In the citing paper
“Deep SORT builds upon the SORT algorithm, which excels in tracking precision and accuracy but struggles with identity switches and occlusions [44].”
From Visual Detection of Traffic Incident through Automatic Monitoring of Vehicle Activities (2024) · cites this paper as [44]
In the citing paper
“These recent but sophisticated tracking algorithms have improved the accuracy of object tracking, but they require heavy computational power.”
From Real-Time Vehicle Classification and Tracking Using a Transfer Learning-Improved Deep Learning Network (2022) · cites this paper as [38]
Citations
Cites
Rich Feature Hierarchies for Accurate Object Detection and Semantic Segmentation
Girshick, R., Donahue, J., Darrell, T., & Malik, J. (2014) · 2014 IEEE Conference on Computer Vision and Pattern Recognition · doi:10.1109/cvpr.2014.81 · Cited by 32,684
He, K., Gkioxari, G., Dollar, P., & Girshick, R. (2017) · 2017 IEEE International Conference on Computer Vision (ICCV) · doi:10.1109/iccv.2017.322 · Cited by 29,137
Object Detection with Discriminatively Trained Part-Based Models
Felzenszwalb, P. F., Girshick, R. B., McAllester, D., & Ramanan, D. (2010) · IEEE Transactions on Pattern Analysis and Machine Intelligence · doi:10.1109/tpami.2009.167 · Cited by 9,853
Simple online and realtime tracking with a deep association metric
Wojke, N., Bewley, A., & Paulus, D. (2017) · 2017 IEEE International Conference on Image Processing (ICIP) · doi:10.1109/icip.2017.8296962 · Cited by 5,286 · cited 3×
“Simple Online and Realtime Tracking with a Deep association metric (Deep SORT) from [20] by N.”
BackgroundMethodScalable Person Re-identification: A Benchmark
Zheng, L., Shen, L., Tian, L., Wang, S., Wang, J., & Tian, Q. (2015) · 2015 IEEE International Conference on Computer Vision (ICCV) · doi:10.1109/iccv.2015.133 · Cited by 4,506 · cited 1×
“The images in this dataset are named by Market1501 person re-identification dataset [21] format.”
MethodSimple online and realtime tracking
Bewley, A., Ge, Z., Ott, L., Ramos, F., & Upcroft, B. (2016) · 2016 IEEE International Conference on Image Processing (ICIP) · doi:10.1109/icip.2016.7533003 · Cited by 4,433 · cited 1×
“The association can be done by spatial information [5, 4] or by both spatial and appearance information [6, 20, 14, 1].”
BackgroundUA-DETRAC: A New Benchmark and Protocol for Multi-Object Detection and Tracking
Wen, L., Du, D., Cai, Z., Lei, Z., Chang, M.-C., Qi, H., Lim, J., Yang, M.-H., & Lyu, S. (2020) · Computer Vision and Image Understanding · doi:10.1016/j.cviu.2020.102907 · Cited by 804 · cited 13×
“Experiments on UA-DETRAC test dataset shows that our proposed extension can achieve promising results by notable margins against state-of-the-art trackers.”
BackgroundMethodInfluentialGlobally-optimal greedy algorithms for tracking a variable number of objects
Pirsiavash, H., Ramanan, D., & Fowlkes, C. C. (2011) · CVPR 2011 · doi:10.1109/cvpr.2011.5995604 · Cited by 790 · cited 1×
“8 GOG[15] CompACT [7] Batch 14.”
High-Speed tracking-by-detection without using image information
Bochinski, E., Eiselein, V., & Sikora, T. (2017) · 2017 14th IEEE International Conference on Advanced Video and Signal Based Surveillance (AVSS) · doi:10.1109/avss.2017.8078516 · Cited by 707 · cited 2×
“in [5], detection results from consequent frames whose intersection-over-union (IOU) is larger than a threshold are associated as a track using greedy algorithm.”
BackgroundMethodChen, L., Ai, H., Zhuang, Z., & Shang, C. (2018) · 2018 IEEE International Conference on Multimedia and Expo (ICME) · doi:10.1109/icme.2018.8486597 · Cited by 530
The Hungarian Method for the Assignment Problem
Kuhn, H. W. (2010) · 50 Years of Integer Programming 1958-2008 · doi:10.1007/978-3-540-68279-0_2 · Cited by 419 · cited 1×
“The new detection results are assigned to the estimated tracks using Hungarian algorithm [13].”
MethodBae, S.-H., & Yoon, K.-J. (2014) · 2014 IEEE Conference on Computer Vision and Pattern Recognition · doi:10.1109/cvpr.2014.159 · Cited by 410 · cited 1×
“CMOT[3] CompACT [7] Online 12.”
Discrete-continuous optimization for multi-target tracking
Andriyenko, A., Schindler, K., & Roth, S. (2012) · 2012 IEEE Conference on Computer Vision and Pattern Recognition · doi:10.1109/cvpr.2012.6247893 · Cited by 336
Deep Cosine Metric Learning for Person Re-identification
Wojke, N., & Bewley, A. (2018) · 2018 IEEE Winter Conference on Applications of Computer Vision (WACV) · doi:10.1109/wacv.2018.00087 · Cited by 329 · cited 3×
“The CNN structure we use is the same as the one in [19], which contains two convolutional layers followed by a max pooling layer and six residual blocks.”
MethodInfluentialLearning Complexity-Aware Cascades for Deep Pedestrian Detection
Cai, Z., Saberian, M., & Vasconcelos, N. (2015) · 2015 IEEE International Conference on Computer Vision (ICCV) · doi:10.1109/iccv.2015.384 · Cited by 326
The Way They Move: Tracking Multiple Targets with Similar Appearance
Dicle, C., Camps, O. I., & Sznaier, M. (2013) · 2013 IEEE International Conference on Computer Vision · doi:10.1109/iccv.2013.286 · Cited by 306
Extending IOU Based Multi-Object Tracking by Visual Information
Bochinski, E., Senst, T., & Sikora, T. (2018) · 2018 15th IEEE International Conference on Advanced Video and Signal Based Surveillance (AVSS) · doi:10.1109/avss.2018.8639144 · Cited by 190 · cited 2×
“[6] extends the IOU tracker to V-IOU tracker by adding visual single-object tracking, which helps IOU tracker to deal with missing detections and reduces the number of ID switches and fragmentations.”
BackgroundMethodMultiple Target Tracking Based on Undirected Hierarchical Relation Hypergraph
Wen, L., Li, W., Yan, J., Lei, Z., Yi, D., & Li, S. Z. (2014) · 2014 IEEE Conference on Computer Vision and Pattern Recognition · doi:10.1109/cvpr.2014.167 · Cited by 170 · cited 1×
“6 H2T[18] CompACT [7] Batch 12.”
UA-DETRAC: A New Benchmark and Protocol for Multi-Object Detection and TrackingPreprint
Wen, L., Du, D., Cai, Z., et al. (2015) · arXiv (Cornell University) · doi:10.48550/arxiv.1511.04136 · Cited by 142 · cited 13×
“Experiments on UA-DETRAC test dataset shows that our proposed extension can achieve promising results by notable margins against state-of-the-art trackers.”
BackgroundMethodInfluentialEvolving boxes for fast vehicle detection
Wang, L., Lu, Y., Wang, H., Zheng, Y., Ye, H., & Xue, X. (2017) · 2017 IEEE International Conference on Multimedia and Expo (ICME) · doi:10.1109/icme.2017.8019461 · Cited by 90 · cited 3×
“Both EB [16] and Mask R-CNN [12] detection methods, which are two stateof-the-art object detection methods we used as the detection input for our tracker, are tested separately.”
MethodMulti-object Tracking Cascade with Multi-Step Data Association and Occlusion Handling
Al-Shakarji, N. M., Bunyak, F., Seetharaman, G., & Palaniappan, K. (2018) · 2018 15th IEEE International Conference on Advanced Video and Signal Based Surveillance (AVSS) · doi:10.1109/avss.2018.8639321 · Cited by 41 · cited 2×
“In [1], a multi-steps data association is proposed, which contains a short-term local association using spatial distance, a global data association with appearance model and a occlusion handling in tracklet level.”
BackgroundMethod
Research works printed but not linked: 1
- chen, 0, Python implementation of the kalman-iou tracker
Cited by
Adnan Yusuf, S., Khan, A., & Souissi, R. (2024) · Transportation Research Interdisciplinary Perspectives · doi:10.1016/j.trip.2023.100980 · Cited by 171
Robust Multi-Drone Multi-Target Tracking to Resolve Target Occlusion: A Benchmark
Liu, Z., Shang, Y., Li, T., Chen, G., Wang, Y., Hu, Q., & Zhu, P. (2023) · IEEE Transactions on Multimedia · doi:10.1109/tmm.2023.3234822 · Cited by 145
Self-citationReal-time multiple object tracking using deep learning methods
Meimetis, D., Daramouskas, I., Perikos, I., & Hatzilygeroudis, I. (2021) · Neural Computing and Applications · doi:10.1007/s00521-021-06391-y · Cited by 122
“For YOLOv3-416 and 608, we are using these anchors: [10, 13], [16, 30], [33, 23], [30, 61], [62, 45], [59, 119], [116, 90], [156, 198] and [373, 326], which correspond to the size of the bounding boxes and are fundamental for the correct training of our CNN.”
BackgroundMethodZhang, S., Abdel-Aty, M. A., Cai, Q., Li, P., & Ugan, J. (2020) · Accident Analysis & Prevention · doi:10.1016/j.aap.2020.105799 · Cited by 91
Neupane, B., Horanont, T., & Aryal, J. (2022) · Sensors · doi:10.3390/s22103813 · Cited by 85
“[38] proposed the use of a method of tracking called DeepSORT with low-confidence track filtering, which reduced the false detection produced by the original DeepSORT algorithm.”
Method
Metadata
- Authors
- Hou, X., Wang, Y., & Chau, L.-P.
- Affiliations
- Nanyang Technological University (SG)
- Venue
- 2019 16th IEEE International Conference on Advanced Video and Signal Based Surveillance (AVSS)
- Publisher
- IEEE
- Published
- 2019-09
- Pages
- 1-6
- OpenAlex
- W2991384465
- Cited by
- 123
- References
- 21
- Citations by year
- 2026: 6 · 2025: 8 · 2024: 26 · 2023: 29 · 2022: 19 · 2021: 25 · 2020: 10
- FWCI
- 4.40
- Type
- conference paper
- Language
- en
- Open access
- closed
- Licence
- Locations
- IEEE International Conference on Advanced Video and Signal Based Surveillance (AVSS) (publishedVersion)
- Topics
- Physical Sciences › Computer Science › Computer Vision and Pattern Recognition › Video Surveillance and Tracking Methods
- Physical Sciences › Engineering › Automotive Engineering › Autonomous Vehicle Technology and Safety
- Physical Sciences › Computer Science › Computer Vision and Pattern Recognition › Advanced Neural Network Applications
- Keywords
- vehicle tracking, multi-target tracking, data association, tracking-by-detection, UA-DETRAC, vehicle re-identification, online 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.