Identity Switch Research Articles

EIoU-distance loss: an automated team-wise player detection and tracking with jersey colour recognition in soccer

The surge in demand for advanced operations in sports video analysis has underscored the crucial role of multiple object tracking. This study addresses the escalating need for efficient and accurate player and referee identification in sports video analysis. The challenge of identity switching among players, especially those with similar appearances, complicates multi-player tracking. Existing algorithms relying on manually labeled data face limitations, particularly with changes in jersey colors. This paper introduces an automated algorithm employing Intersection over Union (IoU) loss and Euclidean Distance (EUD), termed EIoU-Distance Loss, to track players and referees. The method prioritizes identity coherence, aiming to mitigate challenges associated with player and referee recognition. Comprising BackgroundSubtractionMOG2 for player and referee detection and IoU with EUD for connecting nodes across frames, the proposed approach enhances tracking performance, ensuring a clear distinction between different identities. This innovative method addresses critical issues in sports video analysis, offering a robust solution for tracking players and referees in dynamic game scenarios.

An anti-occlusion optimization algorithm for multiple pedestrian tracking.

Frequent occlusion of tracking targets leads to poor performance of tracking algorithms. A common practice in multi-target tracking algorithms is to re-identify the occluded tracking targets, which increases the number of identity switching occurrences. This paper focuses on online multi-object tracking and designs an anti-occlusion, robust association strategy, and feature extraction model. Specifically, the least squares algorithm and the Kalman filter are used to predict the trajectory of the tracking target, while the two-way self-attention mechanism is employed to extract the features of the tracking target, as well as positive and negative samples. After the tracking target is occluded, the association strategy is used to assign the identity information from before the occlusion. The experimental results demonstrate that the algorithm proposed in this paper has achieved excellent tracking performance on the MOT dataset.

Getting action for global economic justice: the micro-foundations of transnational activism

Abstract Generating momentum for activist campaigns on complicated economic issues is difficult, especially in a transnational context. So, how did activists get action on tax justice and create a movement that has changed global tax policy? Drawing on 20 years of para-ethnographic fieldwork with the Tax Justice Network, we suggest that activists initially engaged in ‘identity switching’ tactics to access professional or policy arenas from a footing in one identity, to then switch identities to activate policy shifts. A first-generation leveraged multiple professional identities to access forums, build credibility and introduce a tax lexicon to activists and policymakers. These tactics were not, however, replicable, leading a second generation to concentrate on ‘identity fixing’, including professionalization and a tightening of organizational strategy over access and activation points. Here we theorize identity switching and fixing as underappreciated micro-foundations of transnational activism and demonstrate their importance for global economic justice.

Switching to sustainable products: The role of time, product, and customer characteristics

AbstractAchieving a sustainable economy is one of the primary needs of our time, and sustainable products can play a decisive role in achieving this goal. This article seeks answers about the role of time and product characteristics in the intention to switch to a sustainable product using the example of an everyday product. Drawing from social identity theory, utility maximization and switching theory, the authors propose a conceptual model to examine these questions in a specific market disruption when a product is modified by the same company. The framework focuses on the time‐varying effects of customers' features (brand identification, loyalty, embeddedness) and products' features (perceived value [PV], compliance, switching cost) on switching behavior. The research was carried out using a quasi‐experimental method in three periods. In the first stage, Cox Proportional Hazard Regression was used, and a random‐effects model in the second. Results from longitudinal data of 282 customers during the launch of a new sustainable product show that customer characteristics do not influence the intention to switch. In contrast, the PV and switching costs inhibit the switching behavior, but their effects vary over time. In particular, the effect of the relative PV of the sustainable product on switching will increase over time, but the importance of customer–brand identification also rises. The study has important implications for introducing a sustainable equivalent of an incumbent product of the same brand.

Solving traffic data occlusion problems in computer vision algorithms using DeepSORT and quantum computing

Inaccuracies of traffic sensors during traffic counting and vehicle classification have persisted as transportation agencies have been prompted to calibrate sensors periodically. Detection of multiple objects, heavy occlusions, and similar appearances in congested places are some causes of computer vision model inaccuracies. This paper used the YOLOv5 model for detection and the DeepSORT model for tracking objects. Due to the nature of the reported problem caused by many misses and mismatches, the power of quantum computing with the alternating direction method of multipliers (ADMM) optimizer was leveraged. A basic Kalman filter and the Hungarian algorithm features were used in combination with a quantum optimizer to present robust multiple object tracking (MOT) algorithms. This hybrid combination of the classical and quantum model has fastened learning the occludes during frame matching of tracks and detections by generating minimum quantum cost function value. Comparisons with the existing models indicated a significant increase in the primary MOT metric multiple object tracking accuracy (MOTA) by 16% more than the regular YOLOv5-DeepSORT model when using a quantum optimizer. Also, a 6% multiple object tracking precision (MOTP) increases and a 6% identification metrics (F1) score increase were observed using the quantum optimizer with identity switching reduced from 6 to 4. This model is expected to assist transportation officials in improving the accuracy of traffic counts and vehicle classification and reduce the need for regular computer vision software calibration.

The Development of Various Methods for Object Tracking and Classification in Thermal Videos

Nowadays, thermal cameras play a major role because of its temperature-based photography in many applications such as video surveillance, monitoring electronics/electrical machines, forest monitoring, monitoring babies/adult patients, and suspicious object detection. Tracking pedestrians in thermal video is a major task for such applications. Thermal cameras usually create images based on temperature emitted by the object only and not on the lighting conditions and outdoor environment conditions. But still thermal images have constraints like no texture or colour information, more number of dead pixels, low resolution, and noticeable visual colour patterns in case of any temperature variations. So the challenge in tracking pedestrians in thermal videos is tracking objects/pedestrians throughout the video without an identity switch by overcoming these constraints which may mislead the tracking process. To overcome these constraints, the proposed system uses tri feature matrix (TFM) as an object descriptor which is used to uniquely identify and represent objects in thermal images. TFM is represented in more compact way as a triple matrix. It is a simple and accurate descriptor suitable for tracking objects in thermal video sequences without an identity switch. The proposed Pedestrian tracking system uses most of the advantages of thermal cameras by overcoming challenges in thermal videos effectively based on a novel descriptor TFM. The proposed system is evaluated with various data sets, and the results are analysed using true positive, true negative, false negative, false positive, accuracy, precision, recall, F-score, global identity mismatch (GMME) and track matching error (TME). The performance metrics such as accuracy, precision, recall, F-score, GMME and TME are computed as 99%, 100%, 99%, 99%, 2.3%, and 2.1%, respectively. From the observation, it is found that the performance of proposed TFM-based system is significantly improved. The experimental result shows that the proposed system achieved more accurate tracking compared to the conventional methods

Abstract A109: Single-cell epigenomic analysis reveals an important role of the receptor kinase Ror2 in the erosion of cellular identity during pancreatic carcinogenesis

Abstract Introduction The major driver for pancreatic ductal adenocarcinoma (PDAC) is oncogenic KRAS. However, adult acinar cells, a probable origin of PDAC, are largely refractory to KrasG12D-mediated oncogenic transformation. With the concomitant loss of transcription factors that regulate acinar cell differentiation, such as Pdx1 (Pancreatic and Duodenal Homeobox 1), acinar cells undergo a rapid cell identity switch, known as acinar-to-ductal metaplasia (ADM). How loss of cell identity cooperates with oncogenic Kras to induce pancreatic transformation is largely unclear. Methods To elucidate mechanisms responsible for the accelerated cellular reprogramming in KrasG12D;Pdx1f/f animals, single-cell ATAC-seq (Assay for Transposase-Accessible Chromatin using sequencing) from frozen pancreatic bulk tissue was performed. Chromatin accessibility states were captured at early stages of carcinogenesis and correlated to RNA-seq data. Differentially regulated genes were validated by multiplex RNAscope and immunohistochemistry staining and functionally studied in pancreatic cancer cell lines. Results Single-cell ATAC-seq proved a powerful tool for defining cell-type identity, cellular reprogramming and target genes in early metaplastic transformation of pancreatic tissue. Notably, acinar cells of KrasG12D;Pdx1f/f animals as well as a proportion of metaplastic lesions in both, KrasG12D and KrasG12D;Pdx1f/f mice, showed elevated accessibility and expression of the Ror2 (Receptor Tyrosine Kinase Like Orphan Receptor 2) gene. As a receptor tyrosine kinase, Ror2 controls noncanonical Wnt signaling and other essential signaling pathways, such as PI3K/AKT or Ras-MAPK. Genetic ablation of Ror2 in a mouse model of pancreatic neoplasia resulted in a shift in ADM cell identity, enriching ADM lesions with a senescent duct cell phenotype. In PDAC, ROR2 expression correlates with the more aggressive basal-like subtype. Overexpression of ROR2 in pancreatic cancer cell lines with a classical differentiation induced epithelial-to-mesenchymal transition, characterized by the downregulation of multiple epithelial markers and upregulation of mesenchymal genes. Knockout of ROR2 in pancreatic cancer cells significantly decreased cell proliferation. Conclusions Our in-depth sequencing data revealed that expression of KrasG12D with the concomitant loss of Pdx1 leads to vast alterations of acinar cell identity. We identified the receptor kinase Ror2 as a regulator of pancreatic cancer initiation and driver of pancreatic cancer cell aggressiveness. Citation Format: Simone Benitz, Malak Nasser, Alexander Steep, Jonathan Preall, Ujjwal Mahajan, Ian Loveless, Erick Davis, Hui-Ju Wen, Daniel Long, Michaela Louw, Samuel Zwernik, Donald Rempinski, Jacee Moore, Daniel Salas-Escabillas, Thomas Metzler, Ling Huang, Nina Steele, Ivonne Regel, Filip Bednar, Howard Crawford. Single-cell epigenomic analysis reveals an important role of the receptor kinase Ror2 in the erosion of cellular identity during pancreatic carcinogenesis [abstract]. In: Proceedings of the AACR Special Conference in Cancer Research: Pancreatic Cancer; 2023 Sep 27-30; Boston, Massachusetts. Philadelphia (PA): AACR; Cancer Res 2024;84(2 Suppl):Abstract nr A109.

Multiple Moving Vehicles Tracking Algorithm with Attention Mechanism and Motion Model

With the acceleration of urbanization and the increasing demand for travel, current road traffic is experiencing rapid growth and more complex spatio-temporal logic. Vehicle tracking on roads presents several challenges, including complex scenes with frequent foreground–background transitions, fast and nonlinear vehicle movements, and the presence of numerous unavoidable low-score detection boxes. In this paper, we propose AM-Vehicle-Track, following the proven-effective paradigm of tracking by detection (TBD). At the detection stage, we introduce the lightweight channel block attention mechanism (LCBAM), facilitating the detector to concentrate more on foreground features with limited computational resources. At the tracking stage, we innovatively propose the noise-adaptive extended Kalman filter (NSA-EKF) module to extract vehicles’ motion information while considering the impact of detection confidence on observation noise when dealing with nonlinear motion. Additionally, we borrow the Byte data association method to address unavoidable low-score detection boxes, enabling secondary association to reduce ID switches. We achieve 42.2 MOTA, 51.2 IDF1, and 364 IDs on the test set of VisDrone-MOT with 72 FPS. The experimental results showcase our approach’s highly competitive performance, attaining SOTA tracking performance with a fast speed.

AttMOT: Improving Multiple-Object Tracking by Introducing Auxiliary Pedestrian Attributes.

Multiobject tracking (MOT) is a fundamental problem in computer vision with numerous applications, such as intelligent surveillance and automated driving. Despite the significant progress made in MOT, pedestrian attributes, such as gender, hairstyle, body shape, and clothing features, which contain rich and high-level information, have been less explored. To address this gap, we propose a simple, effective, and generic method to predict pedestrian attributes to support general reidentification (Re-ID) embedding. We first introduce attribute multi-object tracking (AttMOT), a large, highly enriched synthetic dataset for pedestrian tracking, containing over 80k frames and six million pedestrian identity switches (IDs) with different times, weather conditions, and scenarios. To the best of authors' knowledge, AttMOT is the first MOT dataset with semantic attributes. Subsequently, we explore different approaches to fuse Re-ID embedding and pedestrian attributes, including attention mechanisms, which we hope will stimulate the development of attribute-assisted MOT. The proposed method attribute-assisted method (AAM) demonstrates its effectiveness and generality on several representative pedestrian MOT benchmarks, including MOT17 and MOT20, through experiments on the AttMOT dataset. When applied to the state-of-the-art trackers, AAM achieves consistent improvements in multi-object tracking accuracy (MOTA), higher order tracking accuracy (HOTA), association accuracy (AssA), IDs, and IDF1 scores. For instance, on MOT17, the proposed method yields a + 1.1 MOTA, + 1.7 HOTA, and + 1.8 IDF1 improvement when used with FairMOT. To further encourage related research, we release the data and code at https://github.com/HengLan/AttMOT.

An Asymmetric Feature Enhancement Network for Multiple Object Tracking of Unmanned Aerial Vehicle

Multiple object tracking (MOT) in videos captured by unmanned aerial vehicle (UAV) is a fundamental aspect of computer vision. Recently, the one-shot tracking paradigm integrates the detection and re-identification (ReID) tasks, striking a balance between tracking accuracy and inference speed. This paradigm alleviates task conflicts and achieves remarkable results through various feature decoupling methods. However, in challenging scenarios like drone movements, lighting changes and object occlusion, it still encounters issues with detection failures and identity switches. In addition, traditional feature decoupling methods directly employ channel-based attention to decompose the detection and ReID branches, without a meticulous consideration of the specific requirements of each branch. To address the above problems, we introduce an asymmetric feature enhancement network with a global coordinate-aware enhancement (GCAE) module and an embedding feature aggregation (EFA) module, aiming to optimize the two branches independently. On the one hand, we develop the GCAE module for the detection branch, which effectively merges rich semantic information within the feature space to improve detection accuracy. On the other hand, we introduce the EFA module for the ReID branch, which highlights the significance of pixel-level features and acquires discriminative identity embedding through a local feature aggregation strategy. By efficiently incorporating the GCAE and EFA modules into the one-shot tracking pipeline, we present a novel MOT framework, named AsyUAV. Extensive experiments have demonstrated the effectiveness of our proposed AsyUAV. In particular, it achieves a MOTA of 38.3% and IDF1 of 51.7% on VisDrone2019, and a MOTA of 48.0% and IDF1 of 67.5% on UAVDT, outperforming existing state-of-the-art trackers.

INTEGRATING MOTION PRIORS FOR END-TO-END ATTENTION-BASED MULTI-OBJECT TRACKING

Abstract. Recent advancements in multi-object tracking (MOT) have heavily relied on object detection models, with attention-based models like DEtection TRansformer (DETR) demonstrating state-of-the-art capabilities. However, the utilization of attention-based detection models in tracking poses a limitation due to their large parameter count, necessitating substantial training data and powerful hardware for parameter estimation. Ignoring this limitation can lead to a loss of valuable temporal information, resulting in decreased tracking performance and increased identity (ID) switches. To address this challenge, we propose a novel framework that directly incorporates motion priors into the tracking attention layer, enabling an end-to-end solution. Our contributions include: I) a novel approach for integrating motion priors into attention-based multi-object tracking models, and II) a specific realisation of this approach using a Kalman filter with a constant velocity assumption as motion prior. Our method was evaluated on the Multi-Object Tracking dataset MOT17, initial results are reported in the paper. Compared to a baseline model without motion prior, we achieve a reduction in the number of ID switches with the new method.

3D LiDAR Multi-Object Tracking with Short-Term and Long-Term Multi-Level Associations

LiDAR-based Multi-Object Tracking (MOT) is a critical technology employed in various autonomous systems, including self-driving vehicles and autonomous delivery robots. In this paper, a novel LiDAR-based 3D MOT approach is introduced. The proposed method was built upon the Tracking-by-Detection (TbD) paradigm and incorporated multi-level associations that exploit an object’s short-term and long-term relationships with the existing tracks. Specifically, the short-term association leverages the fact that objects do not move much between consecutive frames. In contrast, the long-term association assesses the degree to which a long-term trajectory aligns with current detections. The evaluation of the matching between the current detection and the maintained trajectory was performed using a Graph Convolutional Network (GCN). Furthermore, an inactive track was maintained to address the issue of incorrect ID switching for objects that have been occluded for an extended period. The proposed method was evaluated on the KITTI benchmark MOT tracking dataset and achieved a Higher-Order Tracking Accuracy (HOTA) of 75.65%, marking a 5.66% improvement over the benchmark method AB3DMOT, while also accomplishing the number of ID switches of 39, 74 less than AB3DMOT. These results confirmed the effectiveness of the proposed approach in diverse road environments.

A Multi-Object Tracking Approach Combining Contextual Features and Trajectory Prediction

Aiming to solve the problem of the identity switching of objects with similar appearances in real scenarios, a multi-object tracking approach combining contextual features and trajectory prediction is proposed. This approach integrates the motion and appearance features of objects. The motion features are mainly used for trajectory prediction, and the appearance features are divided into contextual features and individual features, which are mainly used for trajectory matching. In order to accurately distinguish the identities of objects with similar appearances, a context graph is constructed by taking the specified object as the master node and its neighboring objects as the branch nodes. A preprocessing module is applied to exclude unnecessary connections in the graph model based on the speed of the historical trajectory of the object, and to distinguish the features of objects with similar appearances. Feature matching is performed using the Hungarian algorithm, based on the similarity matrix obtained from the features. Post-processing is performed for the temporarily unmatched frames to obtain the final object matching results. The experimental results show that the approach proposed in this paper can achieve the highest MOTA.

Improved UAV-to-Ground Multi-Target Tracking Algorithm Based on StrongSORT.

Unmanned aerial vehicles (UAV) are essential for aerial reconnaissance and monitoring. One of the greatest challenges facing UAVs is vision-based multi-target tracking. Multi-target tracking algorithms that depend on visual data are utilized in a variety of fields. In this study, we present a comprehensive framework for real-time tracking of ground robots in forest and grassland environments. This framework utilizes the YOLOv5n detection algorithm and a multi-target tracking algorithm for monitoring ground robot activities in real-time video streams. We optimized both detection and re-identification networks to enhance real-time target detection. The StrongSORT tracking algorithm was selected carefully to alleviate the loss of tracked objects due to factors like camera jitter, intersecting and overlapping targets, and smaller target sizes. The YOLOv5n algorithm was used to train the dataset, and the StrongSORT tracking algorithm incorporated the best-trained model weights. The algorithm's performance has greatly improved, as demonstrated by experimental results. The number of ID switches (IDSW) has decreased by sixfold, IDF1 has increased by 7.93%, and false positives (FP) have decreased by 30.28%. Additionally, the tracking speed has reached 38 frames per second. These findings validate our algorithm's ability to fulfill real-time tracking requisites on UAV platforms, delivering dependable resolutions for dynamic multi-target tracking on land.

A Computer Vision-Based Algorithm for Detecting Vehicle Yielding to Pedestrians

Computer vision has made remarkable progress in traffic surveillance, but determining whether a motor vehicle yields to pedestrians still requires considerable human effort. This study proposes an automated method for detecting whether a vehicle yields to pedestrians in intelligent transportation systems. The method employs a target-tracking algorithm that uses feature maps and license plate IDs to track the motion of relevant elements in the camera’s field of view. By analyzing the positions of motor vehicles and pedestrians over time, we predict the warning points of pedestrians and hazardous areas in front of vehicles to determine whether the vehicles yield to pedestrians. Extensive experiments are conducted on the MOT16 dataset, real traffic street scene video dataset, and a Unity3D virtual simulation scene dataset combined with SUMO, which demonstrating the superiority of this tracking algorithms. Compared to the current state-of-the-art methods, this method demonstrates significant improvements in processing speed without compromising accuracy. Specifically, this approach substantially outperforms in operational efficiency, thus catering aptly to real-time recognition requirements. This meticulous experimentation and evaluations reveal a commendable reduction in ID switches, enhancing the reliability of violation attributions to the correct vehicles. Such enhancement is crucial in practical urban settings characterized by dynamic interactions and variable conditions. This approach can be applied in various weather, time, and road conditions, achieving high predictive accuracy and interpretability in detecting vehicle–pedestrian interactions. This advanced algorithm illuminates the viable pathways for integrating technological innovation and sustainability, paving the way for more resilient and intelligent urban ecosystems.

A framework for real-time vehicle counting and velocity estimation using deep learning

To better control traffic and promote environmental sustainability, this study proposed a framework to monitor vehicle number and velocity at real time. First, You Only Look Once-v4 (Yolo-v4) algorithm based on deep learning can greatly improve the accuracy of object detection in an image, and trackers, including Sort and Deepsort, resolved the identity switch problem to track efficiently the multiple objects. To that end, this study combined Yolo-v4 with Sort and Deepsort to develop two trajectory models, which are known as YS and YDS, respectively. In addition, different regions of interest (ROI) with different pixel distances (PDs), named ROI-10 and ROI-14, were converted by road marking to calibrate the PD. Finally, a high-resolution benchmark video and two real-time low-resolution videos of highway both were employed to validate this proposed framework. Results show the YDS with ROI-10 achieved 90% accuracy of vehicle counting, when compared to the number of actual vehicles, and this outperformed the YS with ROI-10. However, the YDS with ROI-14 generated relatively good estimates of vehicle velocity. As shown in the real-time low-resolution videos, the YDS with ROI-10 achieved 89.5% and 83.7% accuracy of vehicle counting in Nantun and Daya sites of highway, respectively, and reasonable estimates of vehicle velocity were obtained. In the future, more bus and light truck images could be collected to effectively train the Yolo-v4 and improve the detection of bus and light truck. A better mechanism for precise vehicle velocity estimation and the vehicle detection in different environment conditions should be further investigated.

MBT3D: Deep learning based multi-object tracker for bumblebee 3D flight path estimation.

This work presents the Multi-Bees-Tracker (MBT3D) algorithm, a Python framework implementing a deep association tracker for Tracking-By-Detection, to address the challenging task of tracking flight paths of bumblebees in a social group. While tracking algorithms for bumblebees exist, they often come with intensive restrictions, such as the need for sufficient lighting, high contrast between the animal and background, absence of occlusion, significant user input, etc. Tracking flight paths of bumblebees in a social group is challenging. They suddenly adjust movements and change their appearance during different wing beat states while exhibiting significant similarities in their individual appearance. The MBT3D tracker, developed in this research, is an adaptation of an existing ant tracking algorithm for bumblebee tracking. It incorporates an offline trained appearance descriptor along with a Kalman Filter for appearance and motion matching. Different detector architectures for upstream detections (You Only Look Once (YOLOv5), Faster Region Proposal Convolutional Neural Network (Faster R-CNN), and RetinaNet) are investigated in a comparative study to optimize performance. The detection models were trained on a dataset containing 11359 labeled bumblebee images. YOLOv5 reaches an Average Precision of AP = 53, 8%, Faster R-CNN achieves AP = 45, 3% and RetinaNet AP = 38, 4% on the bumblebee validation dataset, which consists of 1323 labeled bumblebee images. The tracker's appearance model is trained on 144 samples. The tracker (with Faster R-CNN detections) reaches a Multiple Object Tracking Accuracy MOTA = 93, 5% and a Multiple Object Tracking Precision MOTP = 75, 6% on a validation dataset containing 2000 images, competing with state-of-the-art computer vision methods. The framework allows reliable tracking of different bumblebees in the same video stream with rarely occurring identity switches (IDS). MBT3D has much lower IDS than other commonly used algorithms, with one of the lowest false positive rates, competing with state-of-the-art animal tracking algorithms. The developed framework reconstructs the 3-dimensional (3D) flight paths of the bumblebees by triangulation. It also handles and compares two alternative stereo camera pairs if desired.

OTRN-DCN: An optimized transformer-based residual network with deep convolutional network for action recognition and multi-object tracking of adaptive segmentation using soccer sports video

In today’s era, video analysis is immensely involved in recognizing the sport-related movement that has become a significant part of human’s life. The intent of this approach is to know about the player’s activities with prior information of tracking objects. It also analyzes the player potential or capacity to lead the winning team. When the player frequently changes their location, object tracking and action recognition will become a quite challenging task. Over the game, various athletes or different objects are considered to assist the system to easily recognize the respective actions of the player. Most of the previous models have been implemented, yet, it faces such consequences to provide promising performance. To meet the pre-requisite, a new multi-athlete tracking model for action recognition in soccer sports is designed with deep learning approaches. Initially, the multi-object tracking video is offered as the input to pre-processing phase. Here, occlusion and background clutter removal and contrast enhancement techniques are utilized to perform pre-processing in the videos. Then, the pre-processed video is offered to the multi-object tracking phase, where the jersey number is observed during multi-object tracking to avoid the identity switch problem. Then, effective multi-object tracking is performed by adaptive YOLOv5. The parameters presented in the improved adaptive YOLOv5 are tuned by proposing a new algorithm as the Random-based Cheetah Red Deer Algorithm (RCRDA). Next, in the action recognition phase, the tracked object from the video is taken based on the Region of Interest (ROI) that is subjected to an action recognition model named Optimized Transformer-based Residual Network with Deep Convolutional Network (OTRN-DCN). At first, ROI is offered as the input to TRN for attaining the feature vectors. Then, the optimal weighted vector extraction is performed, where the weight is tuned by the developed RCRDA. Finally, the attained optimally weighted vectors are given to the DCN phase for attaining recognized action as output. Hence, the developed multi-object tracking and action recognition model will secure an improved recognition rate than the traditional framework.

Deep-worm-tracker: Deep learning methods for accurate detection and tracking for behavioral studies in C. elegans

Accurate detection and tracking of model organisms such as C. elegans worms remains a fundamental task in behavioral studies. Traditional Machine Learning and Computer Vision methods produce poor detection results and suffer from repeated ID switches during tracking under occlusions and noisy backgrounds. Considering this, we propose Deep-Worm-Tracker, an end-to-end Deep Learning (DL) model, which is a combination of You Only Look Once (YOLOv5) object detection model and Strong Simple Online Real Time Tracking (Strong SORT) tracking backbone that is highly accurate and provides tracking results in real-time inference speeds. Present literature has few solutions to track animals under occlusions and even fewer publicly available large-scale animal re-ID datasets. Thus, we also provide a worm re-ID dataset to minimize worm ID switches, which, to the best of our knowledge, is the first of its kind for C. elegans. We are able to track worms at a mean Average Precision (mAP@0.5) >98% within just 9 min of training time with inference speeds of 9–15 ms for worm detection and on average 27 ms for worm tracking. Our tracking results show that Deep-Worm-Tracker is well suited for ethological studies involving C. elegans.

Enhanced camera-based individual pig detection and tracking for smart pig farms

Negative social interactions are harmful for animal health and welfare. It is increasingly important to employ a continuous and effective monitoring system for detecting and tracking individual animals in large-scale farms. Such a system can provide timely alarms for farmers to intervene when damaging behavior occurs. Deep learning combined with camera-based monitoring is currently arising in agriculture. In this work, deep neural networks are employed to assist individual pig detection and tracking, which enables further analyzing behavior at the individual pig level. First, three state-of-the-art deep learning-based Multi-Object Tracking (MOT) methods are investigated, namely Joint Detection and Embedding (JDE), FairMOT, and YOLOv5s with DeepSORT. All models facilitate automated and continuous individual detection and tracking. Second, weighted-association algorithms are proposed for each MOT method, in order to optimize the object re-identification (re-ID), and improve the individual animal-tracking performance, especially for reducing the number of identity switches. The proposed weighted-association methods are evaluated on a large manually annotated pig dataset, and compared with the state-of-the-art methods. FairMOT with the proposed weighted association achieves the highest IDF1, the least number of identity switches, and the fastest execution rate. YOLOv5s with DeepSORT results in the highest MOTA and MOTP tracking metrics. These methods show high accuracy and robustness for individual pig tracking, and are promising candidates for continuous multi-object tracking for real use in commercial farms.