Ego motion guided particle filter for vehicle tracking in airborne videos

Cao, Xianbin; Chang-cheng, Gao; Lan, Jinhe; Yuan, Yuan; Yan, Pingkun

doi:10.1016/j.neucom.2013.07.014

Cited by 30 publications

(22 citation statements)

References 22 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…In such approaches, discriminative methods using hand-crafted features, such as scale-invariant feature transforms (SIFT), speeded up robust features (SURF), region-based features or edge-based features, are applied for re-identifying vehicles [74][75][76][77][78]. Optical-flow estimation using variational methods [73,79,80], e.g., the Lucas-Kanade method, and correlation-based filters [40], e.g., background-aware correlation filter, are also used for vehicle tracking. Traditional computer vision-based methods, however, generally cannot provide an efficient and reliable feature detector/descriptor for large scale videos of dense urban traffic flows and rarely are tested for multi-vehicle tracking in real-time.…”

Section: Related Workmentioning

confidence: 99%

Orientation- and Scale-Invariant Multi-Vehicle Detection and Tracking from Unmanned Aerial Videos

2019

View full text Add to dashboard Cite

Along with the advancement of light-weight sensing and processing technologies, unmanned aerial vehicles (UAVs) have recently become popular platforms for intelligent traffic monitoring and control. UAV-mounted cameras can capture traffic-flow videos from various perspectives providing a comprehensive insight into road conditions. To analyze the traffic flow from remotely captured videos, a reliable and accurate vehicle detection-and-tracking approach is required. In this paper, we propose a deep-learning framework for vehicle detection and tracking from UAV videos for monitoring traffic flow in complex road structures. This approach is designed to be invariant to significant orientation and scale variations in the videos. The detection procedure is performed by fine-tuning a state-of-the-art object detector, You Only Look Once (YOLOv3), using several custom-labeled traffic datasets. Vehicle tracking is conducted following a tracking-by-detection paradigm, where deep appearance features are used for vehicle re-identification, and Kalman filtering is used for motion estimation. The proposed methodology is tested on a variety of real videos collected by UAVs under various conditions, e.g., in late afternoons with long vehicle shadows, in dawn with vehicles lights being on, over roundabouts and interchange roads where vehicle directions change considerably, and from various viewpoints where vehicles’ appearance undergo substantial perspective distortions. The proposed tracking-by-detection approach performs efficiently at 11 frames per second on color videos of 2720p resolution. Experiments demonstrated that high detection accuracy could be achieved with an average F1-score of 92.1%. Besides, the tracking technique performs accurately, with an average multiple-object tracking accuracy (MOTA) of 81.3%. The proposed approach also addressed the shortcomings of the state-of-the-art in multi-object tracking regarding frequent identity switching, resulting in a total of only one identity switch over every 305 tracked vehicles.

show abstract

Section: Related Workmentioning

confidence: 99%

Orientation- and Scale-Invariant Multi-Vehicle Detection and Tracking from Unmanned Aerial Videos

2019

View full text Add to dashboard Cite

show abstract

“…However, it failed to locate vehicles when the background was highly cluttered. In order to solve this problem, they proposed a novel tracking framework based on the particle filter method [35]. An estimate of the vehicle's motion was incorporated into the particle filter framework to guide particles moving toward the target position.…”

Section: Related Workmentioning

confidence: 99%

A Hierarchical Association Framework for Multi-Object Tracking in Airborne Videos

Chen

Pennisi

et al. 2018

Remote Sensing

View full text Add to dashboard Cite

Multi-Object Tracking (MOT) in airborne videos is a challenging problem due to the uncertain airborne vehicle motion, vibrations of the mounted camera, unreliable detections, changes of size, appearance and motion of the moving objects and occlusions caused by the interaction between moving and static objects in the scene. To deal with these problems, this work proposes a four-stage hierarchical association framework for multiple object tracking in airborne video. The proposed framework combines Data Association-based Tracking (DAT) methods and target tracking using a compressive tracking approach, to robustly track objects in complex airborne surveillance scenes. In each association stage, different sets of tracklets and detections are associated to efficiently handle local tracklet generation, local trajectory construction, global drifting tracklet correction and global fragmented tracklet linking. Experiments with challenging airborne videos show significant tracking improvement compared to existing state-of-the-art methods.

show abstract

“…To conduct a reliable analysis of trajectory data, a framework is needed to efficiently and accurately extract the data. For more than 40 years, researchers have gathered such valuable data by applying the vision-based detection techniques to track vehicles from aerial platforms [5][6][7][8][9] and surveillance cameras mounted at elevated locations [10][11][12].…”

Section: Introductionmentioning

confidence: 99%

“…Although some studies have tried to extract vehicle trajectories from UAV images, they have limited success in improving accuracy due to the use of conventional vehicle detection methods, such as background subtraction, optical flow, and blob analysis. These approaches cannot robustly detect the exact location of vehicles in congested traffic, which results in false tracks in the tracking process [7][8][9]18].…”

Section: Introductionmentioning

confidence: 99%

Extracting Vehicle Trajectories Using Unmanned Aerial Vehicles in Congested Traffic Conditions

Kim

Ham

Kho

et al. 2019

Journal of Advanced Transportation

View full text Add to dashboard Cite

Obtaining the trajectories of all vehicles in congested traffic is essential for analyzing traffic dynamics. To conduct an effective analysis using trajectory data, a framework is needed to efficiently and accurately extract the data. Unfortunately, obtaining accurate trajectories in congested traffic is challenging due to false detections and tracking errors caused by factors in the road environment, such as adjacent vehicles, shadows, road signs, and road facilities. Unmanned aerial vehicles (UAVs), with incorporating machine learning and image processing, can mitigate these difficulties by their ability to hover above the traffic. However, research is lacking regarding the extraction and evaluation of vehicle trajectories in congested traffic. In this study, we propose and compare two learning-based frameworks for detecting vehicles: the aggregated channel feature (ACF), which is based on human-made features, and the faster region-based convolutional neural network (Faster R-CNN), which is based on data-driven features. We extend the detection results to extract vehicle trajectories in congested traffic conditions from UAV images. To remove the errors associated with tracking vehicles, we also develop a postprocessing method based on motion constraints. Then, we conduct detailed performance analyses to confirm the feasibility of the proposed framework on a congested expressway in Korea. The results show that Faster R-CNN outperforms the ACF in images with large objects and in those with small objects if sufficient data are provided. This framework extracts the vehicle trajectories with high precision, making them available for analyzing traffic dynamics based on the training of just a small number of positive samples. The results of this study provide a practical guideline for building a framework to extract vehicles trajectories based on given conditions.

show abstract

Ego motion guided particle filter for vehicle tracking in airborne videos

Cited by 30 publications

References 22 publications

Orientation- and Scale-Invariant Multi-Vehicle Detection and Tracking from Unmanned Aerial Videos

Orientation- and Scale-Invariant Multi-Vehicle Detection and Tracking from Unmanned Aerial Videos

A Hierarchical Association Framework for Multi-Object Tracking in Airborne Videos

Extracting Vehicle Trajectories Using Unmanned Aerial Vehicles in Congested Traffic Conditions

Contact Info

Product

Resources

About