Search

The purpose of this study was to examine the effects of 14-days adenosine 5′-triphosphate (ATP) supplementation (PeakATP®) on reaction time (RT), multiple object tracking speed (MOT), mood and cognition. Twenty adults (22.3 ± 4.4 yrs., 169.9 ± 9.5 cm, 78.7 ± 14.6 kg) completed two experimental trials in a double-blind, counter-balanced, crossover design. Subjects were randomized to either PeakATP® (400 mg) or placebo (PLA) and supplemented for 14-days prior to each trial. During each trial, subjects completed a three-minute all-out test on a cycle ergometer (3MT), with measures of visuomotor RT [Dynavision D2 Proactive (Mode A) and Reactive (Mode B) tasks], MOT (Neurotracker), mood (Profile of Mood States Questionnaire; POMS) and cognition (Automated Neuropsychological Assessment Metrics; ANAM) occurring before (PRE), immediately post (IP) and 60 min post-3MT (60P). Subjects ingested an acute dose of the assigned supplement 30 min prior to completing PRE assessments for each trial. Trials were separated by a 14-day washout period. PeakATP® significantly attenuated declines in hits (p = 0.006, ηp2 = 0.235) and average RT (AvgRT, p = 0.006, ηp2 = 0.236) in Mode A, significantly improved AvgRT (p = 0.039, ηp2 = 0.174) in Mode B, and significantly reduced the total number of misses (p = 0.005, ηp2 = 0.343) in Mode B. No differences between treatments were noted for MOT, POMS or ANAM variables. In conclusion, these results indicate that PeakATP® maintains proactive RT and improves reactive RT following high-intensity sprint exercise suggesting that supplemental ATP may mitigate exercise induced cognitive dysfunction.

Video character tracking problem refers to tracking certain characters of interest in the video and returning the appearing time slots for those characters. Solutions to this problem can be applied in various video-analysis-related areas, such as movie analysis and automatic video clipping. However, there are very few researches investigating this problem and there are no existing relevant benchmark datasets available. In this paper, we design a novel model11 The code for the project is available on https://github.com/zilinghan/victer. to solve this problem by combining a semi-supervised face recognition network and a multi-human tracker. For the face recognition network, we propose a semi-supervised learning method to fully leverage the unlabeled images in the video, thus reducing the required number of labeled face images. Triplet loss is also used during the training to better distinguish among inter-class samples. However, a single face recognition network is insufficient for video character tracking since people do not always show their frontal faces, or sometimes their faces are blocked by some obstacles. Therefore, a multi-human tracker is integrated into the model to address those problems. Additionally, we collect a dataset for the video character tracking problem, Character Face in Video, which can support various experiments for evaluating video character tracker performance. Experiments show that the proposed semi-supervised face recognition model can achieve more than 98.5% recognition accuracy, and our video character tracker can track in near-real-time and achieve 70% ∼ 80% average intersection-over-union tracking accuracy on the dataset.

People "understand" the world via vision. hearing, tactile. and also the past experience. Human experience can be learned through normal learning (we call it explicit knowledge), or subconsciously (we call it implicit knowledge). These experiences learned through normal learning or subconsciously will be encoded and stored in the brain. Using these abundant experience. as a huge database. human beings can effectively process data, even they were unseen beforehand. In this paper. we propose a unified network to encode implicit knowledge and explicit knowledge together, just like the human brain can learn knowledge from normal learning as well as subconsciousness learning. The unified network can generate a unified representation to simultaneously serve various tasks. We can perform kernel space alignment. prediction refinement. and multi-task learning in a convolutional neural network. The results demonstrate that when implicit knowledge is introduced into the neural network, it benefits the performance of all tasks. We further analyze the intplicit representation learnt from the proposed unified network. and it shows great capability on catching the physical meaning of different tasks. The source code of this work is at https:#github.com/WongKinYiu/yolor. [ABSTRACT FROM AUTHOR]

One-shot multiple object tracking (MOT), which learns object detection and identity embedding in a unified network, has attracted increasing attention due to its low complexity and high tracking speed. However, most one-shot trackers ignore that detection and re-identification (ReID) require different representations of features. The inherent difference between these two subtasks leads to optimization contradictions in the training procedure. This issue would result in suboptimal tracking performance. To alleviate this contradiction, we propose a novel dual-path transformation network (DTN) that decouples the shared features into detection-specific and ReID-specific representations. By learning task-specific features, this module satisfies the different requirements of both subtasks. Moreover, we observe that previous trackers generally utilize local information to distinguish targets and ignore global semantic relations, which are crucial for tracking. Therefore, we design a pyramid non-local network (PNN) that allows our network to explore pixel-to-pixel relations with a global receptive field. Meanwhile, PNN considers the scale information to enhance the robustness to scale variations. Extensive experiments conducted on three benchmarks, i.e., MOT16, MOT17, and MOT20, demonstrate the superiority of our tracker, namely DPTrack. The experimental results reveal that DPTrack achieves state-of-the-art performance, e.g., MOTA of 77.1 % and IDF1 of 74.9 % on MOT17. Moreover, DPTrack runs at 14.9FPS, and our lightweight version runs at 26.6FPS with only a slight performance decay. [ABSTRACT FROM AUTHOR]

The preliminary sorting of plastic products is a necessary step to improve the utilization of waste resources. To improve the quality and efficiency of sorting, a plastic detection scheme based on deep learning is proposed in this paper for a waste plastics sorting system based on vision detection. In this scheme, the YOLOX (You Only Look Once) object detection model and the DeepSORT (Deep Simple Online and Realtime Tracking) multiple object tracking algorithm are improved and combined to make them more suitable for plastic sorting. For plastic detection, multiple data augmentations are combined to improve the detection effect, while BN (Batch Normalization) layer fusion and mixed precision inference are adopted to accelerate the model. For plastic tracking, the improved YOLOX is used as a detector, and the tracking effect is further improved by optimizing the deep cosine metric learning and the metric in the matching stage. Based on this, virtual detection lines are set up to filter and extract information to determine the sorted objects. The experimental results show that the scheme proposed in this paper makes full use of vision information to achieve dynamic and real-time detection of plastics. The system is effective and versatile for sorting complex objects. [ABSTRACT FROM AUTHOR]

Multi-object tracking (MOT) has gained significant attention in computer vision due to its wide range of applications. Specifically, detection-based trackers have shown high performance in MOT, but they tend to fail in occlusive scenarios such as the moment when objects overlap or separate. In this paper, we propose a triplet-based MOT network that integrates the tracking information and the visual features of the object. Using a triplet-based image feature, the network can differentiate similar-looking objects, reducing the number of identity switches over a long period. Furthermore, an attention-based re-identification model that focuses on the appearance of objects was introduced to extract the feature vectors from the images to effectively associate the objects. The extensive experimental results demonstrated that the proposed method outperforms existing methods on the ID switch metric and improves the detection performance of the tracking system. [ABSTRACT FROM AUTHOR]

Categorical perception (CP) describes our tendency to perceive the visual world in a categorical manner, suggesting that high-level cognition may affect perception. While most studies are conducted in static visual scenes, Sun and colleagues found CP effects of color in multiple object tracking (MOT). This study used functional magnetic resonance imaging to investigate the neural mechanism behind the categorical effects of color in MOT. Categorical effects were associated with activities in a broad range of brain regions, including both the ventral (V4, middle temporal gyrus) and dorsal pathways (MT + /V5, inferior parietal lobule) of feature processing, as well as frontal regions (middle frontal gyrus, medial superior frontal gyrus). We proposed that these regions are hierarchically organized and responsible for distinct functions. The color-selective V4 encodes color categories, making cross-category colors more discriminable than within-cat- egory colors. Meanwhile, the language and/or semantic regions encode the verbal information of the colors. Both visual and nonvisual codes of color categories then modulate the activities of motion-sensitive MT + areas and frontal areas responsible for attentional processes. [ABSTRACT FROM AUTHOR]

• Created StraTracker framework for strawberry growth monitoring with 90.43% accuracy. • Designed FSA and AKF modules to mitigate light interference and ID switching in MOT. • Introduced a dual-region counting(DC) strategy for fruits at different growth stages. • Constructed StraMOT dataset containing 56,584 frames from 53 videos. Accurately counting fruit in orchards is a critical step for effective digital farming management. However, the variability in fruit size, overlapping shadows, and light interference present significant challenges to applying computer vision during the strawberry growth phase. To address these challenges, we propose StraTracker, a multi-object tracking (MOT) algorithm specifically designed to identify and count strawberries at various growth stages. StraTracker transforms the counting task into a frame-by-frame tracking problem, integrating both motion and appearance features. The algorithm is composed of three key components: a strawberry detector based on YOLOv8n, a feature association module, and a dual-area counting (DC) module. First, the strawberry detector accurately recognizes five growth stages, achieving an average accuracy of 91.93 % at 38.3 FPS. Next, the feature association module, incorporating the Feature Slicing Attention (FSA) and Adaptive Kalman Filtering (AKF) modules, mitigates issues such as light interference, impractical tracking frames, and ID switching (IDs). As a result, StraTracker achieves a Multi-Object Tracking Accuracy (MOTA) of 83.28 % and a Higher-Order Tracking Accuracy (HOTA) of 77.26 %, with only 259 IDs, outperforming existing baseline models. Finally, the DC module categorizes fruit counts based on the unique IDs assigned during tracking. The algorithm's coefficient of determination ( R 2 = 0.91) and GEH of 2.33 indicate a strong correlation between predicted and actual counts. In conclusion, StraTracker offers a promising solution for farmers to optimize planting strategies and develop more precise harvesting plans. [ABSTRACT FROM AUTHOR]

Existing multi-object trackers mainly apply the tracking-by-detection (TBD) paradigm and have achieved remarkable success. However, the mainstream methods execute their detection networks alone, without taking full advantage of the information derived from tracking so that the detection and tracking processes can benefit from each other. In this paper, we achieve strengthened tracking performance in complex scenarios by utilizing the rich temporal information derived from the tracking process to enhance the critical features at the current moment. Specifically, we first propose a critical feature capturing network (CFCN) for extracting receptive field adaptive discriminative features for each frame. Then, we design a temporal-aware feature aggregation module (TFAM), which is used to propagate previous critical features, thus leveraging temporal information to alleviate the detection quality degradation encountered when the visual cues decrease. Extensive experimental comparisons and analyses demonstrate the superiority and effectiveness of the proposed method on the popular and challenging MOT16, MOT17, and MOT20 benchmarks. The experimental results reveal that our tracker achieves state-of-the-art tracking performance, e.g., IDF1 of 75.2% on IDF and MOTA of 80.4% on MOT17. [ABSTRACT FROM AUTHOR]

Object detection and tracking are tasks that humans can perform effortlessly in most environments. Humans can readily recognize individual trees in forests and maintain unique identifiers during occlusion. For computers, on the other hand, this is a complex problem that decades of research have been dedicated to solving. This paper presents a computer vision approach to object detection and tracking tasks in forested environments. We use a state-of-the-art neural network-based detection algorithm to fit bounding boxes around individual tree stems and a simple, efficient, and deterministic multiple object tracking algorithm to maintain unique identities for stems through video frames. We trained the neural network object detector on approximately 3000 ground-truth bounding boxes of ponderosa pine trees. We show that tree stem detection can achieve an average precision of 87% using a Jaccard overlap index of 0.5. We also demonstrate the robustness of the tracking algorithm in occlusion and enter–exit–re-enter scenarios. The presented algorithms can perform object detection and tracking at 49 frames per second on a consumer-grade graphics processing unit. [ABSTRACT FROM AUTHOR]

The corona virus pandemic has introduced limitations which were previously not a cause for concern. Chief among them are wearing face masks in public and constraints on the physical distance between people as an effective measure to reduce the virus spread. Visual surveillance systems, which are common in urban environments and initially commissioned for security surveillance, can be re-purposed to help limit the spread of COVID-19 and prevent future pandemics. In this work, we propose a novel integration technique for real-time pose estimation and multiple human tracking in a pedestrian setting, primarily for social distancing, using CCTV camera footage. Our technique promises a sizeable increase in processing speed and improved detection in very low-resolution scenarios. Using existing surveillance systems, pedestrian pose estimation, tracking and localization for social distancing (PETL4SD) is proposed for measuring social distancing, which combines the output of multiple neural networks aided with fundamental 2D/3D vision techniques. We leverage state-of-the-art object and pose estimation algorithms, combining their strengths, for increase in speed and improvement in detections. These detections are then tracked using a bespoke version of the FASTMOT algorithm. Temporal and analogous estimation techniques are used to deal with occlusions when estimating posture. Projective geometry along with the aforementioned posture tracking is then used to localize the pedestrians. Inter-personal distances are calculated and locally inspected to detect possible violations of the social distancing rules. Furthermore, a “smart violations detector” is employed which estimates if people are together based on their current actions and eliminates false social distancing violations within groups. Finally, distances are intuitively visualized with the right perspective. All implementation is in real time and is performed on Python. Experimental results are provided to validate our proposed method quantitatively and qualitatively on public domain datasets using only a single CCTV camera feed as input. Our results show our technique to outperform the baseline in speed and accuracy in low-resolution scenarios. The code of this work will be made publicly available on GitHub at . [ABSTRACT FROM AUTHOR]

Although the use of a Siamese network is the most popular approach in object tracking, it creates an undesirable trivial solution and requires a large amount of training data reflecting changes in the object's shape in every frame. To solve this problem, in this paper, a self-supervised learning method for multi-object tracking (SSL-MOT) based on a contrastive structure is proposed. Unlike the existing SSL, we adopt a generative adversarial network as a preprocessing step to generate various pose changes of tracking objects. A positive pair composed of the augmented image and pose data is applied to the SSL network to learn an encoder that can generate a non-collapsed output vector. To improve the discrimination power of the encoder output features, we propose an affinity correlation distance, which combines invariance and redundancy terms as a loss function for learning. During the test, because only the dot product between two output vectors of the tracker and detection was used for a data association, the computation time was significantly reduced, and thus real-time online tracking about 12 fps was possible. The proposed method is the first attempt to apply SSL to an online MOT. Experimental results on the MOT16, 17, and 20 challenge datasets proved that the proposed method is a fast and reasonable tracking method that occupies less memory and achieves an excellent tracking performance compared to other state-of-the-art methods. [ABSTRACT FROM AUTHOR]

Recently, there was a lot of researches on real-time detection and tracking algorithms, as the frequent use of surveillance cameras and the expansion of its applications, especially in security and surveillance. However, many challenges have emerged that hinder monitoring systems' work, whether in the detection or tracking stage. We propose a robust new algorithm to detect and track objects from natural scenes captured with real-time cameras to achieve this. This work aims to create a detection and tracking algorithm that is responsive to actual and fundamental changes. This algorithm is characterized by the detection of multiple moving creatures, limited resources, and different challenges. This algorithm combines principal component analysis and deep learning networks to make the most of these two approaches' advantages to achieve an intelligent detection and tracking system that works in real-time. It is done adaptively between the two approaches to enhance performance compared to the existing detection and tracking algorithms. The experimental results showed the new algorithm's effectiveness and efficiency by comparing it with other detection and tracking systems and obtaining good detection and classification accuracy. [ABSTRACT FROM AUTHOR]

The capability of the unmanned aerial vehicle (UAV) to capture highly informative data has expanded its utility in multiple sectors. Surveillance-based UAV applications have highly relied on accurate object tracking. During UAV monitoring, issues such as changes in object appearance and occlusion are common. Tracking the objects under such a scenario is a challenging task. On the other hand, accurate object tracking is quintessential in critical scenarios like security surveillance. In this work, a novel deep learning-based framework for accurate multiple objects tracking with UAV videos is proposed. Tiny-Deeply Supervised Object Detector (Tiny-DSOD) is adopted for accurate object detection. A novel stacked bidirectional-forward LSTM (SBF-LSTM) tracker with spatial and visual features is proposed for object tracking. The spatial and visual features obtained from Tiny-DSOD are trained with the tracker, which predicts object location during tracking. The choice of SBF-LSTM as the tracker enables accurate prediction of object location. Object association is dealt with based on bounding box distance, appearance, and size metrics. With the proposed model, switching object identities is lessened to a greater extent, thereby increasing the tracking accuracy. The proposed methodology outperforms the state-of-the-art methods on UAV videos. [ABSTRACT FROM AUTHOR]

Attention has been shown to modulate the visual evoked potential (VEP) recorded to reversing achromatic patterns. However, the chromatic onset VEP appears to be robust to attentional shifts. Functional magnetic resonance imaging (fMRI) responses to both chromatic and achromatic reversing patterns are also affected by attention. Resolution and comparison of these results is problematic due to differences in presentation mode, stimulus parameters, and the source of the response. Here, we report the results of experiments using comparable perceptual contrasts, pattern reversals, and a co-extensive and highly demanding multiple object tracking (MOT) task while exploring the effects of attentional modulation across both the chromatic (L - M) and (S - (L + M)) and the achromatic visual pathways. Our findings indicate that although achromatic VEPs are modulated by attention, chromatic VEPs are more robust to attentional modulation, even when using comparable stimulus presentation modes and in the presence of a highly demanding distractor task. In addition, we found that the majority of the modulation appears to be from a relative decrease in response due to the distractor task rather than a relative increase in response during heightened attention to the stimulus.

This paper proposes an improvement to the multi-object tracking system framework based on the image inputs. By analyzing the role and performance of each block in the original multi-objects tracking system, the blocks of the original system are reconstructed to enhance the efficiency and yield a faster processing speed suiting the real-time applications. In the proposed method, the first two parts of the multi-object tracking system are merged into a single neural network designed for object detection and feature extraction. A new object association judgment method and JDE inspired prediction head are included in order to achieve a better and an outstanding association effect resulting in the overall improvement of the original system by 45.2%. The enhanced method is aimed at the application of smart roadside units and uses fixed-viewpoint image input to achieve multi-object tracking on embedded platforms. The proposed method is implemented on the NVIDIA Jetson AGX Xavier embedded platform. The NVIDIA TensorRT software development kit is used to accelerate the neural network. The overall performance of the proposed system yields better efficiency compared to that of the original SDE design and the overall computing performance achieve up to 14–26 images per second, making it ideal for the real-time smart roadside unit applications.

To shed light on the processes driving cell migration, a systematic analysis of the cell behavior is required. Since the manual analysis of hundreds or even thousands of cells is infeasible, automated approaches for cell segmentation and tracking are needed. While for the task of cell segmentation deep learning has become the standard, there are few approaches for simultaneous cell segmentation and tracking using deep learning. Here, we present EmbedTrack, a single convolutional neural network for simultaneous cell segmentation and tracking which predicts human comprehensible embeddings. As embeddings, offsets of cell pixels to their cell center and bandwidths are learned which are processed in a subsequent clustering step to generate an instance segmentation and link the segmented instances over time. We benchmark our approach on nine 2D data sets from the Cell Tracking Challenge, where our approach performs on seven out of nine data sets within the top 3 contestants including three top 1 performances. The source code is publicly available at https://git.scc.kit.edu/kit-loe-ge/embedtrack.

This study was conducted to examine differences in visual attention according to sports type and sex. In total, 132 participants [open-skill sport athletes (basketball players), closed-skill sport athletes (swimmers), and non-athletes; n = 22 men and 22 women each] aged 19–24 years performed a multiple object tracking (MOT) task, which is a well-established paradigm for the assessment of visual attention. Visual tracking accuracy was affected by the sport type (p

Multiple object tracking (MOT), as a core technology for environment perception in autonomous driving, has attracted attention from researchers. Combing the advantages of batch global optimization, we present a novel online MOT framework for autonomous driving, consisting of feature extraction and data association on a temporal window. In the feature extraction stage, we design a three-channel appearance feature extraction network based on metric learning by using ResNet50 as the backbone network and the triplet loss function and employ a Kalman Filter with a constant acceleration motion model to optimize and predict the object bounding box information, so as to obtain reliable and discriminative object representation features. For data association, to reduce the ID switches, the min-cost flow of global association is introduced within the temporal window composed of consecutive multi-frame images. The trajectories within the temporal window are divided into two categories, active trajectories and inactive trajectories, and the appearance, motion affinities between each category of trajectories, and detections are calculated, respectively. Based on this, a sparse affinity network is constructed, and the data association is achieved using the min-cost flow problem of the network. Qualitative experimental results on KITTI MOT public benchmark dataset and real-world campus scenario sequences validate the effectiveness and robustness of our method. Compared with the homogeneous, vision-based MOT methods, quantitative experimental results demonstrate that our method has competitive advantages in terms of higher order tracking accuracy, association accuracy, and ID switches.

Keeping track of the location of multiple moving objects is one of the well documented functions of visual attention. However, the mechanism of attentional selection that supports such continuous tracking is unclear. In particular, it has been proposed that target selection in early visual cortex occurs in parallel, with tracking errors arising because of attentional limitations at later processing stages. Here, we examine whether, instead, total attentional capacity for enhancement of early visual processing of tracked targets is shared between all attended stimuli. If the magnitude of attentional facilitation of multiple tracked targets was a key limiting factor of tracking ability, then one should expect it to drop systematically with increasing set-size of tracked targets. Human observers (male and female) were instructed to track two, four, or six moving objects among a pool of identical distractors. Steady-state visual evoked potentials (SSVEPs) recorded during the tracking period revealed that the processing of tracked targets was consistently amplified compared with the processing of the distractors. The magnitude of this amplification decreased with increasing set size, and at lateral occipital electrodes it closely followed inverse proportionality to the number of tracked items, suggesting that limited attentional resources must be shared among the tracked stimuli. Accordingly, the magnitude of attentional facilitation predicted the behavioral outcome at the end of the trial. Together, these findings demonstrate that the limitations of multiple object tracking (MOT) across set-sizes stem from the limitations of top-down selective attention already at the early stages of visual processing. [ABSTRACT FROM AUTHOR]

Previous studies of multiple object tracking suggested that spatiotemporal features (e.g., speed, direction, location) and surface features (e.g., color, shape, size) can guide perceptual grouping. However, it is still unclear how target-distractor distinctiveness and target-background similarity affect grouping effects in attentive tracking. To address these two questions, three experiments have been carried out in the current study. In Experiment 1, we manipulated the target-distractor distinctiveness and found that tracking performance logarithmically improved when the target-distractor distinctiveness linearly increased. In Experiment 2a and 2b, we varied the target-background similarity and found that too high or too low target-background similarity damaged the tracking performance, while only the middle target-background similarity resulted in the best tracking performance. These findings reveal that not only target-distractor distinctiveness but also target-background similarity plays a vital role in guiding the attention of perceptual grouping in attentive tracking. The guidance induced by target-distractor distinctiveness is constrained by the spatial resolution, while the guidance induced by target-background similarity is constrained by the efficiency of object segmentation. Additionally, our results showed that tracking capacity varied with the target-distractor distinctiveness and the target-background similarity, even though the number of targets being tracked was fixed. It suggests that there may be a trade-off between the difficulty of tracking and the number of targets that can be tracked. Thus, tracking capacity is more likely to be limited by the flexible attention resources rather than the number of fixed slots. [ABSTRACT FROM AUTHOR]

The process of object tracking involves consistently identifying each instance across frames depending on initial set of object detection(s). Moreover, in multiple object tracking (MOT), the process through tracking-by-detection paradigm consists of performing two common steps consecutively, which are detection and data association. In MOT, it is targeted to associate detections across frames by localizing and identifying all objects of interest. MOT algorithms further keep tracking even the most challenging issues such as revisiting the same view, missing detections, occlusion and temporarily unseen objects, same-appearance objects coexisting in the same frame occur. Hence, re-identification (re-id) appears to be the most powerful tool for assigning the correct identities to each individual instance when aforementioned issues arise. In this work, we propose a similarity-based person re-id framework, called SAT, using a Siamese neural network via shared weights. Once detections are obtained from the backbone SAT applies a Siamese feature extraction model and then we introduce a similarity array for assessing tracklet(s) and detection(s). We examine the performance of SAT on several benchmarks with extensive experiments and statistical tests, where we improve the current state-of-the-art according to commonly used performance metrics with higher accuracy, less ID switches, less false positive and negative rates. [ABSTRACT FROM AUTHOR]

The recent advancement in autonomous robotics is directed toward designing a reliable system that can detect and track multiple objects in the surrounding environment for navigation and guidance purposes. This paper aims to survey the recent development in this area and present the latest trends that tackle the challenges of multiple object tracking, such as heavy occlusion, dynamic background, and illumination changes. Our research includes Multiple Object Tracking (MOT) methods incorporating the multiple inputs that can be perceived from sensors such as cameras and Light Detection and Ranging (LIDAR). In addition, a summary of the tracking techniques, such as data association and occlusion handling, is detailed to define the general framework that the literature employs. We also provide an overview of the metrics and the most common benchmark datasets, including Karlsruhe Institute of Technology and Toyota Technological Institute (KITTI), MOTChallenges, and University at Albany DEtection and TRACking (UA-DETRAC), that are used to train and evaluate the performance of MOT. At the end of this paper, we discuss the results gathered from the articles that introduced the methods. Based on our analysis, deep learning has introduced significant value to the MOT techniques in recent research, resulting in high accuracy while maintaining real-time processing. [ABSTRACT FROM AUTHOR]

Multiple object tracking technology has been widely applied in video analysis, signal processing and other fields. The object detector's performance determines the tracking accuracy and speed, in the "tracking by detection" mode that modern multiple object tracking systems usually follow. A dedicated cyclic detector is proposed to improve the tracking performance, which uses the characteristics of high similarity between video frames. The candidate frame is selected by considering object position information in the previous frame, and variation score map of current frame relative to the previous frame, which solves the problem of large parameters and calculations caused by region proposal network in traditional two-stage target detector. At the same time, dedicated cyclic detector integrating the object appearance feature extraction branch can further reduce the overall running time of multiple object tracking system. Dedicated cyclic detector and other state-of-the-art detectors are respectively applied to multiple object tracking system. Experimental results prove that dedicated cyclic detector can improve the tracking speed while ensuring the tracking accuracy. [ABSTRACT FROM AUTHOR]

Background: Although an extensive body of literature is trying to verify the acute effects of exercise, findings are highly contradictory due to many different study protocols. The number of studies using an intermittent exercise (IE) protocol is limited, especially with regard to comparison across the life span. We examined whether the effects of a HIIE protocol on performance in a perceptual-cognitive task (NeuroTracker® (NT)) differed between children, young adults, and older adults to address this gap. Methods: A total of 36 participants participated in the present study: 12 children (CH, 6 females, 9.83 ± 1.19 years), 12 young adults (YA, 6 females, 23.5 ± 3.55 years), and 12 older adults (OA, 4 females, 66.92 ± 4.08 years). The IE treadmill protocol used in the present study consisted of eleven 30- second intervals at 90% VO2max, interspersed with 2-minute active recovery periods at 50% VO2max. Before and during this exercise protocol, three series of the NeuroTracker® task were performed after 5, 15, and 25 minutes. Results: We observed a significant main effect time and a significant main effect group regarding absolute NT scores and progression during IE. YA had significantly higher absolute NT scores than CH and OA. The normalized perceptual-cognitive task progression was observed in OA and YA but not in CH. YA, in particular, showed progression in the NT performance during IE. Conclusions: The present study confirmed previous findings on age-related differences in NT performance. Based on these findings, the effects of different exercise protocols (e.g., continuous vs. intermittent) seem to be a worthwhile subject for future investigations. Normalized speed thresholds should best capture improvement differences between groups to compare results across studies better, as pre-test values are taken as the baseline. [ABSTRACT FROM AUTHOR]

Background: The ability to track multiple objects plays a key role in team ball sports actions. However, there is a lack of research focused on identifying multiple object tracking (MOT) performance under rapid, dynamic and ecologically valid conditions. Therefore, we aimed to assess the effects of manipulating postural stability on MOT performance. Methods: Nineteen team sports players (soccer, basketball, handball) and sixteen sedentary individuals performed the MOT task under three levels of postural stability (high, medium, and low). For the MOT task, participants had to track three out of eight balls for 10 s, and the object speed was adjusted following a staircase procedure. For postural stability manipulation, participants performed three identical protocols (randomized order) of the MOT task while standing on an unstable platform, using the training module of the Biodex Balance System SD at levels 12 (high-stability), eight (medium-stability), and four (low-stability). Results: We found that the ability to track moving targets is dependent on the balance stability conditions (F2,66 = 8.7, p < 0.001, ² = 0.09), with the disturbance of postural stability having a negative effect on MOT performance. Moreover, when compared to sedentary individuals, team sports players showed better MOT scores for the high-stability and the medium-stability conditions (corrected p-value = 0.008, Cohen's d = 0.96 and corrected p-value = 0.009, Cohen's d = 0.94; respectively) whereas no differences were observed for the more unstable conditions (lowstability) between-groups. Conclusions: The ability to track moving targets is sensitive to the level of postural stability, with the disturbance of balance having a negative effect on MOT performance. Our results suggest that expertise in team sports training is transferred to non-specific sport domains, as shown by the better performance exhibited by team sports players in comparison to sedentary individuals. This study provides novel insights into the link between individual's ability to track multiple moving objects and postural control in team sports players and sedentary individuals. [ABSTRACT FROM AUTHOR]

Background The ability to track multiple objects plays a key role in team ball sports actions. However, there is a lack of research focused on identifying multiple object tracking (MOT) performance under rapid, dynamic and ecologically valid conditions. Therefore, we aimed to assess the effects of manipulating postural stability on MOT performance. Methods Nineteen team sports players (soccer, basketball, handball) and sixteen sedentary individuals performed the MOT task under three levels of postural stability (high, medium, and low). For the MOT task, participants had to track three out of eight balls for 10 s, and the object speed was adjusted following a staircase procedure. For postural stability manipulation, participants performed three identical protocols (randomized order) of the MOT task while standing on an unstable platform, using the training module of the Biodex Balance System SD at levels 12 (high-stability), eight (medium-stability), and four (low-stability). Results We found that the ability to track moving targets is dependent on the balance stability conditions (F2,66 = 8.7, p < 0.001, η² = 0.09), with the disturbance of postural stability having a negative effect on MOT performance. Moreover, when compared to sedentary individuals, team sports players showed better MOT scores for the high-stability and the medium-stability conditions (corrected p-value = 0.008, Cohen’s d = 0.96 and corrected p-value = 0.009, Cohen’s d = 0.94; respectively) whereas no differences were observed for the more unstable conditions (low-stability) between-groups. Conclusions The ability to track moving targets is sensitive to the level of postural stability, with the disturbance of balance having a negative effect on MOT performance. Our results suggest that expertise in team sports training is transferred to non-specific sport domains, as shown by the better performance exhibited by team sports players in comparison to sedentary individuals. This study provides novel insights into the link between individual’s ability to track multiple moving objects and postural control in team sports players and sedentary individuals.

Multi-object tracking in unmanned aerial vehicle (UAV) videos is a critical visual perception task with numerous applications. However, existing multi-object tracking methods, when directly applied to UAV scenarios, face significant challenges in maintaining robust tracking due to factors such as motion blur and small object sizes. Additionally, existing UAV methods tend to underutilize crucial information from the temporal and spatial dimensions. To address these issues, on the one hand, we propose a temporal feature aggregation module (TFAM), which effectively combines temporal contexts to obtain rich feature response maps in dynamic motion scenes to enhance the detection capability of the proposed tracker. On the other hand, we introduce a topology-integrated embedding module (TIEM) that captures the topological relationships between objects and their surrounding environment globally and sparsely, thereby integrating spatial layout information. The proposed TIEM significantly enhances the discriminative power of object embedding features, resulting in more precise data association. By integrating these two carefully designed modules into a one-stage online MOT system, we construct a robust UAV tracker. Compared to the baseline approach, the proposed model demonstrates significant improvements in MOTA on two UAV multi-object tracking benchmarks, namely VisDrone2019 and UAVDT. Specifically, the proposed model achieves a 2.2% improvement in MOTA on the VisDrone2019 benchmark and a 2.5% improvement on the UAVDT benchmark.

Generic multiple object tracking aims to recover the trajectories for generic moving objects of the same category. This task relies on the ability of effectively extracting representative features of the target objects. To this end, we propose a novel prototype learning based model, PLGMOT, that can explore the template features of an exemplar object and extend to more objects to acquire their prototype. Their prototype features can be continuously updated during the video, in favor of generalization to all the target objects with different appearances. More importantly, on the public benchmark GMOT-40, our method achieves more than 14% advantage over the state-of-the-art methods, with less than 0.5% of the training data that is not even completely annotated in the form of bounding boxes, thanks to our proposed point-to-box label refinement training algorithm and hierarchical motion-aware association algorithm. • A prototype learning based generic multi-object detector. • A point-to-box label refinement training algorithm. • A hierarchical motion-aware association algorithm for tracking. • Extensive experiments demonstrate the effectiveness of the proposed method. [ABSTRACT FROM AUTHOR]

Multiple Object Tracking (MOT) has potential applications in construction site safety man- agement, particularly for the individualized assessment of scaffold workers' safety status. However, its accuracy can be degraded due to inconsistent detection results and insuffi- cient object association capabilities. This paper addresses these challenges by proposing an object tracking method, named as the Shallow Cascaded Buffered Intersection over Union (Shallow C-BIoU). This non-machine learning-based tracker employs a color hash- ing technique to enhance tracking performance. Employing rigorous evaluation metrics, the experimental results underscore the efficacy of the proposed method. Compared to state-of- the-art algorithms, the Shallow C-BIoU method improved association performance by 7.56%, reducing 52.44% of falsely assigned tracking IDs. Consequently, this paper contributes to the development of reliable object trackers, thereby advancing monitoring technologies for construction management purposes. • Introduced Shallow C-BIoU for enhanced worker tracking. • Achieved 52.44% reduction in false tracking IDs. • Improved association performance by 7.56% using color hashing. • Managed occlusions efficiently in scaffolding scenes. • Offers unique datasets for multi object tracking research advancement. [ABSTRACT FROM AUTHOR]

Multiple object tracking (MOT) in unmanned aerial vehicle (UAV) videos is a fundamental task and can be applied in many fields. MOT consists of two critical procedures, i.e., object detection and re-identification (ReID). One-shot MOT, which incorporates detection and ReID in a unified network, has gained attention due to its fast inference speed. It significantly reduces the computational overhead by making two subtasks share features. However, most existing one-shot trackers struggle to achieve robust tracking in UAV videos. We observe that the essential difference between detection and ReID leads to an optimization contradiction within one-shot networks. To alleviate this contradiction, we propose a novel feature decoupling network (FDN) to convert shared features into detection-specific and ReID-specific representations. The FDN searches for characteristics and commonalities between the two tasks to synergize detection and ReID. In addition, existing one-shot trackers struggle to locate small targets in UAV videos. Therefore, we design a pyramid transformer encoder (PTE) to enrich the semantic information of the resulting detection-specific representations. By learning scale-aware fine-grained features, the PTE empowers our tracker to locate targets in UAV videos accurately. Extensive experiments on VisDrone2021 and UAVDT benchmarks demonstrate that our tracker achieves state-of-the-art tracking performance. [ABSTRACT FROM AUTHOR]

Previous studies on perceptual grouping found that people can use spatiotemporal and featural information to group spatially separated rigid objects into a unit while tracking moving objects. However, few studies have tested the role of objects' self‐motion information in perceptual grouping, although it is of great significance to the motion perception in the three‐dimensional space. In natural environments, objects always move in translation and rotation at the same time. The self‐rotation of the objects seriously destroys objects' rigidity and topology, creates conflicting movement signals and results in crowding effects. Thus, this study sought to examine the specific role played by self‐rotation information on grouping spatially separated non‐rigid objects through a modified multiple object tracking (MOT) paradigm with self‐rotating objects. Experiment 1 found that people could use self‐rotation information to group spatially separated non‐rigid objects, even though this information was deleterious for attentive tracking and irrelevant to the task requirements, and people seemed to use it strategically rather than automatically. Experiment 2 provided stronger evidence that this grouping advantage did come from the self‐rotation per se rather than surface‐level cues arising from self‐rotation (e.g. similar 2D motion signals and common shapes). Experiment 3 changed the stimuli to more natural 3D cubes to strengthen the impression of self‐rotation and again found that self‐rotation improved grouping. Finally, Experiment 4 demonstrated that grouping by self‐rotation and grouping by changing shape were statistically comparable but additive, suggesting that they were two different sources of the object information. Thus, grouping by self‐rotation mainly benefited from the perceptual differences in motion flow fields rather than in deformation. Overall, this study is the first attempt to identify self‐motion as a new feature that people can use to group objects in dynamic scenes and shed light on debates about what entities/units we group and what kinds of information about a target we process while tracking objects. [ABSTRACT FROM AUTHOR]