UA-DETRAC: A New Benchmark and Protocol for Multi-Object Detection and Tracking

Wen, Longyin; Du, Dawei; Cai, Zhaowei; Lei, Zhen; Chang, Ming-Ching; Qi, Honggang; Lim, Jongwoo; Yang, Ming–Hsuan; Lyu, Siwei

doi:10.48550/arxiv.1511.04136

Cited by 30 publications

(43 citation statements)

References 70 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The multi-object tracking (MOT) is a critical task for autonomous driving, as it needs to perform object detection as well as object tracking in a video. A large array of datasets have been created focusing on driving scenarios, for example, KITTI tracking [10], MOTChallenge [31], UA-DETRAC [42], PathTrack [30], and Pose-Track [1]. None of these datasets provide segmentation masks for the annotated objects and thus do not depict pixellevel representations and complex interactions like MOTS Figure 2.…”

Section: Related Workmentioning

confidence: 99%

DG-Labeler and DGL-MOTS Dataset: Boost the Autonomous Driving Perception

Cui¹,

Cao²,

Xie³

et al. 2021

Preprint

View full text Add to dashboard Cite

show abstract

Section: Related Workmentioning

confidence: 99%

DG-Labeler and DGL-MOTS Dataset: Boost the Autonomous Driving Perception

Cui¹,

Cao²,

Xie³

et al. 2021

Preprint

View full text Add to dashboard Cite

show abstract

“…The UA-DETRAC Benchmark Suite contains 10 hours of video containing traffic sequences divided into 60 training and 40 testing videos. The training and test data contain an average of 7.1 and 12.0 objects per frame, respectively [48]. LBT-extended trackers are evaluated according to the PR-metrics defined in [48] which evaluate tracking performance at varying levels of detection confidence levels.…”

Section: Ua-detracmentioning

confidence: 99%

“…The training and test data contain an average of 7.1 and 12.0 objects per frame, respectively [48]. LBT-extended trackers are evaluated according to the PR-metrics defined in [48] which evaluate tracking performance at varying levels of detection confidence levels. Each tracker is evaluated on the training dataset with varying numbers of frames between detection d. Tracking is performed at d = 0, 1, 3, 7, 15 and 31 frames.…”

Section: Ua-detracmentioning

confidence: 99%

See 1 more Smart Citation

Localization-Based Tracking

Gloudemans¹,

Work²

2021

Preprint

View full text Add to dashboard Cite

End-to-end production of object tracklets from high resolution video in real-time and with high accuracy remains a challenging problem due to the cost of object detection on each frame. In this work we present Localization-based Tracking (LBT), an extension to any tracker that follows the tracking by detection or joint detection and tracking paradigms. Localization-based Tracking focuses only on regions likely to contain objects to boost detection speed and avoid matching errors. We evaluate LBT as an extension to two example trackers (KIOU and SORT) on the UA-DETRAC and MOT20 datasets. LBT-extended trackers outperform all other reported algorithms in terms of PR-MOTA, PR-MOTP, and mostly tracked objects on the UA-DETRAC benchmark, establishing a new state-of-the art. Relative to tracking by detection with KIOU, LBT-extended KIOU achieves a 25% higher frame-rate and is 1.1% more accurate in terms of PR-MOTA on the UA-DETRAC dataset. LBT-extended SORT achieves a 62% speedup and a 3.2% increase in PR-MOTA on the UA-DETRAC dataset. On MOT20, LBT-extended KIOU has a 50% higher frame-rate than tracking by detection and is 0.4% more accurate in terms of MOTA. As of submission time, our LBT-extended KIOU tracker places 10th overall on the MOT20 benchmark.

show abstract

“…Existing vision-based multiple object tracking (MOT) datasets and benchmarks, such as KITTI [30], MOTChallenge [22], UA-DETRAC [107] and NuScenes [16] have been instrumental for advancing and monitoring the progress of MOT methods in well-controlled settings. Here, the central task is the detection and tracking of multiple objects from a predefined closed set of classes, such as cars and pedestrians.…”

Section: Introductionmentioning

confidence: 99%

Opening up Open-World Tracking

Liu¹,

Zulfikar²,

Luiten³

et al. 2021

Preprint

View full text Add to dashboard Cite

In this paper, we propose and study Open-World Tracking (OWT). Open-world tracking goes beyond current multiobject tracking benchmarks and methods which focus on tracking object classes that belong to a predefined closedset of frequently observed object classes. In OWT, we relax this assumption: we may encounter objects at inference time that were not labeled for training. The main contribution of this paper is the formalization of the OWT task, along with an evaluation protocol and metric (Open-World Tracking Accuracy, OWTA), which decomposes into two intuitive terms, one for measuring recall, and another for measuring track association accuracy. This allows us to perform a rigorous evaluation of several different baselines that follow design patterns proposed in the multi-object tracking community. Further we show that our Open-World Tracking Baseline, while performing well in the OWT setting, also achieves near state-of-the-art results on traditional closedworld benchmarks, without any adjustments or tuning. We believe that this paper is an initial step towards studying multi-object tracking in the open world, a task of crucial importance for future intelligent agents that will need to understand, react to, and learn from, an infinite variety of objects that can appear in an open world.

show abstract

UA-DETRAC: A New Benchmark and Protocol for Multi-Object Detection and Tracking

Cited by 30 publications

References 70 publications

DG-Labeler and DGL-MOTS Dataset: Boost the Autonomous Driving Perception

DG-Labeler and DGL-MOTS Dataset: Boost the Autonomous Driving Perception

Localization-Based Tracking

Opening up Open-World Tracking

Contact Info

Product

Resources

About