Explore the vast range of titles available.

Home-range estimation is an important application of animal tracking data that is frequently complicated by autocorrelation, sampling irregularity, and small effective sample sizes. We introduce a novel, optimal weighting method that accounts for temporal sampling bias in autocorrelated tracking data. This method corrects for irregular and missing data, such that oversampled times are downweighted and undersampled times are upweighted to minimize error in the home-range estimate. We also introduce computationally efficient algorithms that make this method feasible with large data sets. Generally speaking, there are three situations where weight optimization improves the accuracy of home-range estimates: with marine data, where the sampling schedule is highly irregular, with duty cycled data, where the sampling schedule changes during the observation period, and when a small number of home-range crossings are observed, making the beginning and end times more independent and informative than the intermediate times. Using both simulated data and empirical examples including reef manta ray, Mongolian gazelle, and African buffalo, optimal weighting is shown to reduce the error and increase the spatial resolution of home-range estimates. With a conveniently packaged and computationally efficient software implementation, this method broadens the array of data sets with which accurate space-use assessments can be made.

The study of animal movement and behavior is being revolutionized by technology, such as satellite tags and harmonic radar, that allows us to track the movements of individual animals. However, our ability to analyze and model such data has lagged behind the sophisticated collection methods. We review problems with current methods and suggest a more powerful and flexible approach, state-space modeling, and we illustrate how these models can be posed in a meta-analytic framework so that information from individual trajectories may be combined optimally. State-space models enable us to deal with the complexity of modeling animals interacting with their environment but, unlike other methods, they allow simultaneous estimation of measurement error and process noise that are inherent in animal-trajectory data. A Bayesian framework allows us to incorporate important prior information when available and also allows meta-analytic techniques to be incorporated in a straightforward fashion. Meta-analysis enables both individual and broader-level inference from observations of multiple individual pathways. Our approach is powerful because it allows researchers to test hypotheses regarding animal movement, to connect theoretical models to data, and to use modern likelihood-based estimation techniques, all under a single statistical framework.

In-situ visual observations of marine organisms is crucial to developing behavioural understandings and their relations to their surrounding ecosystem. Typically, these observations are collected via divers, tags, and remotely-operated or human-piloted vehicles. Recently, however, autonomous underwater vehicles equipped with cameras and embedded computers with GPU capabilities are being developed for a variety of applications, and in particular, can be used to supplement these existing data collection mechanisms where human operation or tags are more difficult. Existing approaches have focused on using fully-supervised tracking methods, but labelled data for many underwater species are severely lacking. Semi-supervised trackers may offer alternative tracking solutions because they require less data than fully-supervised counterparts. However, because there are not existing realistic underwater tracking datasets, the performance of semi-supervised tracking algorithms in the marine domain is not well understood. To better evaluate their performance and utility, in this paper we provide (1) a novel dataset specific to marine animals located at http://warp.whoi.edu/vmat/, (2) an evaluation of state-of-the-art semi-supervised algorithms in the context of underwater animal tracking, and (3) an evaluation of real-world performance through demonstrations using a semi-supervised algorithm on-board an autonomous underwater vehicle to track marine animals in the wild.

Effective ocean management and the conservation of highly migratory species depend on resolving the overlap between animal movements and distributions, and fishing effort. However, this information is lacking at a global scale. Here we show, using a big-data approach that combines satellite-tracked movements of pelagic sharks and global fishing fleets, that 24% of the mean monthly space used by sharks falls under the footprint of pelagic longline fisheries. Space-use hotspots of commercially valuable sharks and of internationally protected species had the highest overlap with longlines (up to 76% and 64%, respectively), and were also associated with significant increases in fishing effort. We conclude that pelagic sharks have limited spatial refuge from current levels of fishing effort in marine areas beyond national jurisdictions (the high seas). Our results demonstrate an urgent need for conservation and management measures at high-seas hotspots of shark space use, and highlight the potential of simultaneous satellite surveillance of megafauna and fishers as a tool for near-real-time, dynamic management. A global dataset of the satellite-tracked movements of pelagic sharks and fishing fleets show that sharks—and, in particular, commercially important species—have limited spatial refuge from fishing effort.

With the development of computer vision technology, many advanced computer vision methods have been successfully applied to animal detection, tracking, recognition and behavior analysis, which is of great help to ecological protection, biodiversity conservation and environmental protection. As existing datasets applied to target tracking contain various kinds of common objects, but rarely focus on wild animals, this paper proposes the first benchmark, named Wild Animal Tracking Benchmark (WATB), to encourage further progress of research and applications of visual object tracking. WATB contains more than 203,000 frames and 206 video sequences, and covers different kinds of animals from land, sea and sky. The average length of the videos is over 980 frames. Each video is manually labelled with thirteen challenge attributes including illumination variation, rotation, deformation, and so on. In the dataset, all frames are annotated with axis-aligned bounding boxes. To reveal the performance of these existing tracking algorithms and provide baseline results for future research on wild animal tracking, we benchmark a total of 38 state-of-the-art trackers and rank them according to tracking accuracy. Evaluation results demonstrate that the trackers based on deep networks perform much better than other trackers like correlation filters. Another finding on the basis of the evaluation results is that wild animals tracking is still a big challenge in computer vision community. The benchmark WATB and evaluation results are released on the project website https://w-1995.github.io/.

Researchers have long attempted to follow animals as they move through their environment. Until relatively recently, however, such efforts were limited to short distances and times in species large enough to carry large batteries and transmitters. New technologies have opened up new frontiers in animal tracking remote data collection. Hussey et al. review the unique directions such efforts have taken for marine systems, while Kays et al. review recent advances for terrestrial species. We have entered a new era of animal ecology, where animals act as both subjects and samplers of their environments. Science , this issue 10.1126/science.1255642 , 10.1126/science.aaa2478 Moving animals connect our world, spreading pollen, seeds, nutrients, and parasites as they go about the their daily lives. Recent integration of high-resolution Global Positioning System and other sensors into miniaturized tracking tags has dramatically improved our ability to describe animal movement. This has created opportunities and challenges that parallel big data transformations in other fields and has rapidly advanced animal ecology and physiology. New analytical approaches, combined with remotely sensed or modeled environmental information, have opened up a host of new questions on the causes of movement and its consequences for individuals, populations, and ecosystems. Simultaneous tracking of multiple animals is leading to new insights on species interactions and, scaled up, may enable distributed monitoring of both animals and our changing environment.

Electronic tags are significantly improving our understanding of aquatic animal behavior and are emerging as key sources of information for conservation and management practices. Future aquatic integrative biology and ecology studies will increasingly rely on data from electronic tagging. Continued advances in tracking hardware and software are needed to provide the knowledge required by managers and policymakers to address the challenges posed by the world’s changing aquatic ecosystems. We foresee multiplatform tracking systems for simultaneously monitoring the position, activity, and physiology of animals and the environment through which they are moving. Improved data collection will be accompanied by greater data accessibility and analytical tools for processing data, enabled by new infrastructure and cyberinfrastructure. To operationalize advances and facilitate integration into policy, there must be parallel developments in the accessibility of education and training, as well as solutions to key governance and legal issues.

Advanced general visual object tracking models have been drastically developed with the access of large annotated datasets and progressive network architectures. However, a general tracker always suffers domain shift when directly adopting to specific testing scenarios. In this paper, we dedicate to addressing the animal tracking problem by proposing a spatio-temporal inference module and a coarse-to-fine tracking strategy. In terms of tracking animals, non-rigid deformation is a typical challenge. Therefore, we particularly design a novel transformer-based inference structure where the changing animal state is transmitted across continuous frames. By explicitly transmitting the appearance variations, this spatio-temporal module enables adaptive target learning, boosting the animal tracking performance compared to the fixed template matching approaches. Besides, considering the altered contours of animals in different frames, we propose to perform coarse-to-fine tracking to obtain a fine-grained animal bounding box with a dedicated distribution-aware regression module. The coarse tracking phase focuses on distinguishing the target against potential distractors in the background. While the fine-grained tracking phase aims at accurately regressing the final animal bounding box. To facilitate animal tracking evaluation, we captured and annotated 145 video sequences with 20 categories from the zoo, forming a new test set for animal tracking, coined ZOO145. We also collected a dataset, AnimalSOT, with 162 video sequences from existing tracking test benchmarks. The experimental performance on animal tracking datasets, MoCA, ZOO145, and AnimalSOT, demonstrate the merit of the proposed approach against advanced general tracking approaches, providing a baseline for future animal tracking studies.

Markerless methods for animal posture tracking have been rapidly developing recently, but frameworks and benchmarks for tracking large animal groups in 3D are still lacking. To overcome this gap in the literature, we present 3D-MuPPET, a framework to estimate and track 3D poses of up to 10 pigeons at interactive speed using multiple camera views. We train a pose estimator to infer 2D keypoints and bounding boxes of multiple pigeons, then triangulate the keypoints to 3D. For identity matching of individuals in all views, we first dynamically match 2D detections to global identities in the first frame, then use a 2D tracker to maintain IDs across views in subsequent frames. We achieve comparable accuracy to a state of the art 3D pose estimator in terms of median error and Percentage of Correct Keypoints. Additionally, we benchmark the inference speed of 3D-MuPPET, with up to 9.45 fps in 2D and 1.89 fps in 3D, and perform quantitative tracking evaluation, which yields encouraging results. Finally, we showcase two novel applications for 3D-MuPPET. First, we train a model with data of single pigeons and achieve comparable results in 2D and 3D posture estimation for up to 5 pigeons. Second, we show that 3D-MuPPET also works in outdoors without additional annotations from natural environments. Both use cases simplify the domain shift to new species and environments, largely reducing annotation effort needed for 3D posture tracking. To the best of our knowledge we are the first to present a framework for 2D/3D animal posture and trajectory tracking that works in both indoor and outdoor environments for up to 10 individuals. We hope that the framework can open up new opportunities in studying animal collective behaviour and encourages further developments in 3D multi-animal posture tracking.

The advent of miniaturized biologging devices has provided ecologists with unprecedented opportunities to record animal movement across scales, and led to the collection of ever‐increasing quantities of tracking data. In parallel, sophisticated tools have been developed to process, visualize and analyse tracking data; however, many of these tools have proliferated in isolation, making it challenging for users to select the most appropriate method for the question in hand. Indeed, within the r software alone, we listed 58 packages created to deal with tracking data or ‘tracking packages’. Here, we reviewed and described each tracking package based on a workflow centred around tracking data (i.e. spatio‐temporal locations (x, y, t)), broken down into three stages: pre‐processing, post‐processing and analysis, the latter consisting of data visualization, track description, path reconstruction, behavioural pattern identification, space use characterization, trajectory simulation and others. Supporting documentation is key to render a package accessible for users. Based on a user survey, we reviewed the quality of packages' documentation and identified 11 packages with good or excellent documentation. Links between packages were assessed through a network graph analysis. Although a large group of packages showed some degree of connectivity (either depending on functions or suggesting the use of another tracking package), one third of the packages worked in isolation, reflecting a fragmentation in the r movement‐ecology programming community. Finally, we provide recommendations for users when choosing packages, and for developers to maximize the usefulness of their contribution and strengthen the links within the programming community. The increased use of biologging devices has propelled the development of methods and software tools for analyzing tracking data. This work reviews 58 r packages for movement, acts as a road map for movement ecologists and offers recommendations for package developers from a user perspective.