Explore the vast range of titles available.

Shadows are widely used in the tracking of moving targets by video synthetic aperture radar (video SAR). However, they always appear in groups in video SAR images. In such cases, track effects produced by existing single-target tracking methods are no longer satisfactory. To this end, an effective way to obtain the capability of multiple target tracking (MTT) is in urgent demand. Note that tracking by detection (TBD) for MTT in optical images has achieved great success. However, TBD cannot be utilized in video SAR MTT directly. The reasons for this is that shadows of moving target are quite different from in video SAR image than optical images as they are time-varying and their pixel sizes are small. The aforementioned characteristics make shadows in video SAR images hard to detect in the process of TBD and lead to numerous matching errors in the data association process, which greatly affects the final tracking performance. Aiming at the above two problems, in this paper, we propose a multiple target tracking method based on TBD and the Siamese network. Specifically, to improve the detection accuracy, the multi-scale Faster-RCNN is first proposed to detect the shadows of moving targets. Meanwhile, dimension clusters are used to accelerate the convergence speed of the model in the training process as well as to obtain better network weights. Then, SiamNet is proposed for data association to reduce matching errors. Finally, we apply a Kalman filter to update the tracking results. The experimental results on two real video SAR datasets demonstrate that the proposed method outperforms other state-of-art methods, and the ablation experiment verifies the effectiveness of multi-scale Faster-RCNN and SimaNet.

Multi-rotor drones have expanded their range of applications, one of which being search and rescue (SAR) missions using infrared thermal imaging. This paper addresses thermal target tracking with track segment association (TSA) for SAR missions. Three types of associations including TSA are developed with an interacting multiple model (IMM) approach. During multiple-target tracking, tracks are initialized, maintained, and terminated. There are three different associations in track maintenance: measurement–track association, track–track association for tracks that exist at the same time (track association and fusion), and track–track association for tracks that exist at separate times (TSA). Measurement–track association selects the statistically nearest measurement and updates the track with the measurement through the IMM filter. Track association and fusion fuses redundant tracks for the same target that are spatially separated. TSA connects tracks that have become broken and separated over time. This process is accomplished through the selection of candidate track pairs, backward IMM filtering, association testing, and an assignment rule. In the experiments, a drone was equipped with an infrared thermal imaging camera, and two thermal videos were captured of three people in a non-visible environment. These three hikers were located close together and occluded by each other or other obstacles in the mountains. The drone was allowed to move arbitrarily. The tracking results were evaluated by the average total track life, average mean track life, and average track purity. The track segment association improved the average mean track life of each video by 99.8% and 250%, respectively.

Target localization in a wireless sensor network (WSN) has received more and more attention in recent years, and has promoted many new applications due to the low cost, low bandwidth, low energy consumption, and collision avoidance mechanism. How to provide accurate location information has always been a hot research topic in 5G/B5G application scenarios. In this paper, the path loss information or received signal strength (RSS) of the received signal is considered in a WSN for the extended Kalman filter (EKF) to realize trajectory tracking of multiple targets, and the tracked targets are then localized through multiple sensors. Moreover, since there may be several objects or clutter interference in the communication environment, in order to reduce the impact of interference, we consider the probabilistic data association filter (PDAF) or probability hypothesis density filter (PHDF) to improve the tracking performance. Each sensor sends the received distance estimation information to the fusion center (FC), which calculates the optimal position for each target. Through simulation results, the proposed weighted least squares (WLS) trilateration method in this paper can effectively improve the average root mean squared error (RMSE) performance as sensors are evenly distributed around the tracking trajectories.

In this paper, a novel frame-by-frame data association algorithm based on intuitionistic fuzzy sets is proposed for online visual multiple target tracking. In the proposed algorithm, the association costs between targets and measurements are replaced by the intuitionistic fuzzy membership degrees which are obtained by a modified intuitionistic fuzzy c-means clustering algorithm. In addition, in order to mine useful information from the uncertain measurements, a new intuitionistic index is defined and the intuitionistic fuzzy point operator is applied to extract valuable information from the intuitionistic index. Experiments with challenging public datasets demonstrate that the proposed visual tracking algorithm improves tracking performance compared to other algorithms.

In order to simultaneously improve system performance and resource utilization of distributed multiple-input multiple-output (MIMO) radar systems, a joint resource allocation method is proposed to address the velocity estimation problem for multiple targets tracking in this paper. The paper focuses to improve the tracking performance for key targets using the remaining resources when the general targets have obtained resources to reach to tracking requirements. Firstly, a criterion minimizing the velocity estimation mean square error (MSE) for a key target is considered. Restricted by limited and relatively sufficient system resources and given velocity estimation requirements for general targets, a joint resource allocation optimization model with transmitters, receivers, transmitted power, and signal time is established. We propose a suboptimal method to approximately solve this problem. The method separates the optimization into three steps, where each step transforms the corresponding mixed-Boolean optimization problem into a second-order cone programming (SOCP) problem by convex relaxation. Finally, the approximately optimal solution can be obtained by cyclic minimization method. Extensive simulations indicate that compared with other methods, the proposed joint method can achieve the lowest velocity estimation MSE with the fewest transmitters. Meanwhile, limited by the given velocity estimation MSE, the proposed method can focus on the key target and achieve the whole velocity estimation error minimization while a greater flexibility for target tracking number can be obtained. Moreover, random experiments can further validate and evaluate the proposed method’s effectiveness and traceability with the given scenario.

Multiple targets tracking is a major issue in the intelligent applications. Numerous methods have been presented for the multiple targets tracking to capture the targets trajectory in a video sequence in order to increase intelligence and reduce human error. In this paper, a method is proposed based on combining the Extended Kalman Filter (EKF) and Particle Swarm Optimization (PSO) to construct an intermediate tracker and track targets more accurately. The EKF solves targets collision problem, and PSO reduces the covariance of measured noise. Finally, the Joint Probabilistic Data Association (JPDA) filter is used to reduce the number of multiple hypotheses and create a one-to-one correspondence between targets and measurements. To detect targets, frames subtracting along with background modeling and canny edge detector are used. To reduce running time of the proposed method, number of video frames per second (fps) is reduced from 30 to 10 and the sampling rate is also reduced. Despite of this reduction, simulation reults of the proposed method show the multiple targets tracking with 98% accuracy at an acceptable running time compared to the similar methods. In addition, by using the proposed method, the number of assignment states is reduced in the targets tracking process. Overall, the proposed method not only can be used in the intelligent applications, but also in the video compression applications as well. Article Highlights Multiple targets tracker with 98% accuracy based on combining the EKF and PSO Execution speed increasing by reduction of the video fps from 30 to 10 Ambiguous measurements reduction using JPDA in the targets identification assignment process

Multi-object tracking (MOT) has been notoriously difficult to evaluate. Previous metrics overemphasize the importance of either detection or association. To address this, we present a novel MOT evaluation metric, higher order tracking accuracy (HOTA), which explicitly balances the effect of performing accurate detection, association and localization into a single unified metric for comparing trackers. HOTA decomposes into a family of sub-metrics which are able to evaluate each of five basic error types separately, which enables clear analysis of tracking performance. We evaluate the effectiveness of HOTA on the MOTChallenge benchmark, and show that it is able to capture important aspects of MOT performance not previously taken into account by established metrics. Furthermore, we show HOTA scores better align with human visual evaluation of tracking performance.

Multi-object tracking (MOT) is an important problem in computer vision which has a wide range of applications. Formulating MOT as multi-task learning of object detection and re-ID in a single network is appealing since it allows joint optimization of the two tasks and enjoys high computation efficiency. However, we find that the two tasks tend to compete with each other which need to be carefully addressed. In particular, previous works usually treat re-ID as a secondary task whose accuracy is heavily affected by the primary detection task. As a result, the network is biased to the primary detection task which is not fair to the re-ID task. To solve the problem, we present a simple yet effective approach termed as FairMOT based on the anchor-free object detection architecture CenterNet. Note that it is not a naive combination of CenterNet and re-ID. Instead, we present a bunch of detailed designs which are critical to achieve good tracking results by thorough empirical studies. The resulting approach achieves high accuracy for both detection and tracking. The approach outperforms the state-of-the-art methods by a large margin on several public datasets. The source code and pre-trained models are released at https://github.com/ifzhang/FairMOT.

For the problem that traditional data association algorithms tend to coalesce neighboring tracks for multiple close targets tracking application in dense clutter, measurements adaptive censor (MAC) method to Set JPDA (SJPDA) algorithm was introduced in this paper, then the proposed the MACSJPDA algorithm of target tracking discards several data associations with small probability and accelerates the convergence speed of the SJPDA algorithm. The algorithm can achieve better effects of multiple targets tracking by multiple sensors in wireless sensor networks. Monte Carlo simulation revealed that estimation effect of the MACSJPDA algorithm is much smoother, and it needs less run time than SJPDA algorithm for handling closely spaced and crossing targets, in the meanwhile the mean optimal sub-pattern assignment (MOSPA) deviation of the MACSJPDA algorithm is also smaller.

With the increasing popularity of Unmanned Aerial Vehicles (UAVs) in computer vision-related applications, intelligent UAV video analysis has recently attracted the attention of an increasing number of researchers. To facilitate research in the UAV field, this paper presents a UAV dataset with 100 videos featuring approximately 2700 vehicles recorded under unconstrained conditions and 840k manually annotated bounding boxes. These UAV videos were recorded in complex real-world scenarios and pose significant new challenges, such as complex scenes, high density, small objects, and large camera motion, to the existing object detection and tracking methods. These challenges have encouraged us to define a benchmark for three fundamental computer vision tasks, namely, object detection, single object tracking (SOT) and multiple object tracking (MOT), on our UAV dataset. Specifically, our UAV benchmark facilitates evaluation and detailed analysis of state-of-the-art detection and tracking methods on the proposed UAV dataset. Furthermore, we propose a novel approach based on the so-called Context-aware Multi-task Siamese Network (CMSN) model that explores new cues in UAV videos by judging the consistency degree between objects and contexts and that can be used for SOT and MOT. The experimental results demonstrate that our model could make tracking results more robust in both SOT and MOT, showing that the current tracking and detection methods have limitations in dealing with the proposed UAV benchmark and that further research is indeed needed.