Computer vision analysis of camera trap video footage is essential for wildlife conservation, as captured behaviours offer some of the earliest indicators of changes in population health. Recently, several high-impact animal behaviour datasets and methods have been introduced to encourage their use; however, the role of behaviour-correlated background information and its significant effect on out-of-distribution generalisation remain unexplored. In response, we present the PanAf-FGBG dataset, featuring 21 hours of wild chimpanzee behaviours, recorded at 389 individual camera locations. Uniquely, it pairs every video with a chimpanzee (referred to as a foreground video) with a corresponding background video (with no chimpanzee) from the same camera location. We present two views of the dataset: one with overlapping camera locations and one with disjoint locations. This setup enables, for the first time, direct evaluation of in-distribution and out-of-distribution conditions, and for the impact of backgrounds on behaviour recognition models to be quantified. All clips have rich behavioural annotations and metadata, including unique camera IDs. Additionally, we establish several baselines and present a highly effective latent-space normalisation technique that boosts out-of-distribution performance by +5.42% mAP for convolutional and +3.75% mAP for transformer-based models. Finally, we provide an in-depth analysis on the role of backgrounds in out-of-distribution behaviour recognition, including the so far unexplored impact of background durations. (i.e., the count of background frames within foreground videos). The dataset is available at https://obrookes.github.io/panaf-fgbg.github.io/.

Visual as well as genetic biometrics are routinely employed to identify species and individuals in biological applications. However, no attempts have been made in this domain to computationally enhance visual classification of rare classes with little image data via genetics. In this paper, we thus propose aligned visual-genetic learning as a new application domain with the aim to implicitly encode cross-modality associations for improved performance. We demonstrate for the first time that such alignment can be achieved via deep embedding models and that the approach is directly applicable to boosting long-tailed recognition (LTR), particularly for rare species. We experimentally demonstrate the efficacy of the concept via application to microscopic imagery of 30k+ planktic foraminifer shells across 32 species when used together with independent genetic data samples. Most importantly for practitioners, we show that visual-genetic alignment can significantly benefit visual-only recognition of the rarest species. Technically, we pre-train a visual ResNet50 deep learning model using triplet loss formulations to create an initial embedding space. We re-structure this space based on genetic anchors embedded via a Sequence Graph Transform (SGT) and linked to visual data by cross-domain cosine alignment. We show that an LTR approach improves the state-of-the-art across all benchmarks and that adding our visual-genetic alignment improves per-class and particularly rare tail class benchmarks significantly further. Overall, visual-genetic LTR training raises rare per-class accuracy from 37.4% to benchmark-beating 59.7%. We conclude that visual-genetic alignment can be a highly effective tool for complementing visual biological data containing rare classes. The concept proposed may serve as an important future tool for integrating genetics and imageomics towards a more complete scientific representation of taxonomic spaces and life itself. Code, weights, and data splits are published for full reproducibility.

Abstract We present the PanAf20K dataset, the largest and most diverse open-access annotated video dataset of great apes in their natural environment. It comprises more than 7 million frames across $$\sim $$ ∼ 20,000 camera trap videos of chimpanzees and gorillas collected at 18 field sites in tropical Africa as part of the Pan African Programme: The Cultured Chimpanzee. The footage is accompanied by a rich set of annotations and benchmarks making it suitable for training and testing a variety of challenging and ecologically important computer vision tasks including ape detection and behaviour recognition. Furthering AI analysis of camera trap information is critical given the International Union for Conservation of Nature now lists all species in the great ape family as either Endangered or Critically Endangered. We hope the dataset can form a solid basis for engagement of the AI community to improve performance, efficiency, and result interpretation in order to support assessments of great ape presence, abundance, distribution, and behaviour and thereby aid conservation efforts. The dataset and code are available from the project website: PanAf20K

This paper presents an investigation into long-tail video recognition. We demonstrate that, unlike naturally-collected video datasets and existing long-tail image benchmarks, current video benchmarks fall short on multiple long-tailed properties. Most critically, they lack few-shot classes in their tails. In response, we propose new video benchmarks that better assess long-tail recognition, by sampling subsets from two datasets: SSv2 and VideoLT. We then propose a method, Long-Tail Mixed Reconstruction (LMR), which reduces overfitting to instances from few-shot classes by reconstructing them as weighted combinations of samples from head classes. LMR then employs label mixing to learn robust decision boundaries. It achieves state-of-the-art average class accuracy on EPIC-KITCHENS and the proposed SSv2-LT and VideoLT-LT. Benchmarks and code at: github.com/tobyperrett/lmr

This paper presents a deep learning framework for medical video segmentation. Convolution neural network (CNN) and transformer-based methods have achieved great milestones in medical image segmentation tasks due to their incredible semantic feature encoding and global information comprehension abilities. However, most existing approaches ignore a salient aspect of medical video data - the temporal dimension. Our proposed framework explicitly extracts features from neighbouring frames across the temporal dimension and incorporates them with a temporal feature blender, which then tokenises the high-level spatio-temporal feature to form a strong global feature encoded via a Swin Transformer. The final segmentation results are produced via a UNet-like encoder-decoder architecture. Our model outperforms other approaches by a significant margin and improves the segmentation benchmarks on the VFSS2022 dataset, achieving a dice coefficient of 0.8986 and 0.8186 for the two datasets tested. Our studies also show the efficacy of the temporal feature blending scheme and cross-dataset transferability of learned capabilities. Code and models are available at https://github.com/SimonZeng7108/Video-SwinUNet.

Abstract We propose a novel end-to-end curriculum learning approach for sparsely labelled animal datasets leveraging large volumes of unlabelled data to improve supervised species detectors. We exemplify the method in detail on the task of finding great apes in camera trap footage taken in challenging real-world jungle environments. In contrast to previous semi-supervised methods, our approach adjusts learning parameters dynamically over time and gradually improves detection quality by steering training towards virtuous self-reinforcement. To achieve this, we propose integrating pseudo-labelling with curriculum learning policies and show how learning collapse can be avoided. We discuss theoretical arguments, ablations, and significant performance improvements against various state-of-the-art systems when evaluating on the Extended PanAfrican Dataset holding approx. 1.8M frames. We also demonstrate our method can outperform supervised baselines with significant margins on sparse label versions of other animal datasets such as Bees and Snapshot Serengeti. We note that performance advantages are strongest for smaller labelled ratios common in ecological applications. Finally, we show that our approach achieves competitive benchmarks for generic object detection in MS-COCO and PASCAL-VOC indicating wider applicability of the dynamic learning concepts introduced. We publish all relevant source code, network weights, and data access details for full reproducibility.

We propose Video-TransUNet, a deep architecture for instance segmentation in medical CT videos constructed by integrating temporal feature blending into the TransUNet deep learning framework. In particular, our approach amalgamates strong frame representation via a ResNet CNN backbone, multi-frame feature blending via a Temporal Context Module (TCM), non-local attention via a Vision Transformer, and reconstructive capabilities for multiple targets via a UNet-based convolutional-deconvolutional architecture with multiple heads. We show that this new network design can significantly outperform other state-of-the-art systems when tested on the segmentation of bolus and pharynx/larynx in Videofluoroscopic Swallowing Study (VFSS) CT sequences. On our VFSS2022 dataset it achieves a dice coefficient of 0.8796 and an average surface distance of 1.0379 pixels. Note that tracking the pharyngeal bolus accurately is a particularly important application in clinical practice since it constitutes the primary method for diagnostics of swallowing impairment. Our findings suggest that the proposed model can indeed enhance the TransUNet architecture via exploiting temporal information and improving segmentation performance by a significant margin. We publish key source code, network weights, and ground truth annotations for simplified performance reproduction.

We propose a novel approach to few-shot action recognition, finding temporally-corresponding frame tuples between the query and videos in the support set. Distinct from previous few-shot works, we construct class prototypes using the CrossTransformer attention mechanism to observe relevant sub-sequences of all support videos, rather than using class averages or single best matches. Video representations are formed from ordered tuples of varying numbers of frames, which allows sub-sequences of actions at different speeds and temporal offsets to be compared.1Our proposed Temporal-Relational CrossTransformers (TRX) achieve state-of-the-art results on few-shot splits of Kinetics, Something-Something V2 (SSv2), HMDB51 and UCF101. Importantly, our method outperforms prior work on SSv2 by a wide margin (12%) due to the its ability to model temporal relations. A detailed ablation showcases the importance of matching to multiple support set videos and learning higher-order relational CrossTransformers.

This paper describes a computationally-enhanced M100 UAV platform with an onboard deep learning inference system for integrated computer vision and navigation. The system is able to autonomously find and visually identify by coat pattern individual Holstein Friesian cattle in freely moving herds. We propose an approach that utilises three deep convolutional neural network architectures running live onboard the aircraft: (1) a YOLOv2-based species detector, (2) a dual-stream deep network delivering exploratory agency, and (3) an InceptionV3-based biometric long-term recurrent convolutional network for individual animal identification. We evaluate the performance of each of the components offline, and also online via real-world field tests comprising 147 minutes of autonomous low altitude flight in a farm environment over a dispersed herd of 17 heifer dairy cows. We report error-free identification performance on this online experiment. The presented proof-of-concept system is the first of its kind. It represents a practical step towards autonomous biometric identification of individual animals from the air in open pasture environments for tag-less AI support in farming and ecology.

This paper discusses deep learning for solving static and dynamic search and recovery tasks - such as the retrieval of all instances of actively moving targets - based on partial-view Unmanned Aerial Vehicle (UAV)-like sensing. In particular, we demonstrate that abstracted tactic and strategic explorational agency can be implemented effectively via a single deep network that optimises in unity: the mapping of sensory inputs and positional history towards navigational actions. We propose a dual-stream classification paradigm that integrates one Convolutional Neural Network (CNN) for sensory processing with a second one for interpreting an evolving longterm map memory. In order to learn effective search behaviours given agent location and agent-centric sensory inputs, we train this design against 400k+ optimal navigational decision samples from each set of static and dynamic evolutions for different multi-target behaviour classes. We quantify recovery performance across an extensive range of scenarios; including probabilistic placement and dynamics, as well as fully random target walks and herd-inspired behaviours. Detailed results comparisons show that our design can outperform naive, independent stream and off-the-shelf DRQN solutions. We conclude that the proposed dual-stream architecture can provide a unified, rationally motivated and effective architecture for solving online search tasks in dynamic, multi-target environments. With this paper we publish3key source code and associated models.

Deriving a person's energy expenditure accurately forms the foundation for tracking physical activity levels across many health and lifestyle monitoring tasks. In this study, the authors present a method for estimating calorific expenditure from combined visual and accelerometer sensors by way of an RGB‐Depth camera and a wearable inertial sensor. The proposed individual‐independent framework fuses information from both modalities which leads to improved estimates beyond the accuracy of single modality and manual metabolic equivalents of task (MET) lookup table based methods. For evaluation, the authors introduce a new dataset called SPHERE_RGBD + Inertial_calorie , for which visual and inertial data are simultaneously obtained with indirect calorimetry ground truth measurements based on gas exchange. Experiments show that the fusion of visual and inertial data reduces the estimation error by 8 and 18% compared with the use of visual only and inertial sensor only, respectively, and by 33% compared with a MET‐based approach. The authors conclude from their results that the proposed approach is suitable for home monitoring in a controlled environment.

Multiple human tracking (MHT) is a fundamental task in many computer vision applications. Appearance‐based approaches, primarily formulated on RGB data, are constrained and affected by problems arising from occlusions and/or illumination variations. In recent years, the arrival of cheap RGB‐depth devices has led to many new approaches to MHT, and many of these integrate colour and depth cues to improve each and every stage of the process. In this survey, the authors present the common processing pipeline of these methods and review their methodology based (a) on how they implement this pipeline and (b) on what role depth plays within each stage of it. They identify and introduce existing, publicly available, benchmark datasets and software resources that fuse colour and depth data for MHT. Finally, they present a brief comparative evaluation of the performance of those works that have applied their methods to these datasets.

This paper discusses the automated visual identification of individual great white sharks from dorsal fin imagery. We propose a computer vision photo ID system and report recognition results over a database of thousands of unconstrained fin images. To the best of our knowledge this line of work establishes the first fully automated contour-based visual ID system in the field of animal biometrics. The approach put forward appreciates shark fins as textureless, flexible and partially occluded objects with an individually characteristic shape. In order to recover animal identities from an image we first introduce an open contour stroke model, which extends multi-scale region segmentation to achieve robust fin detection. Secondly, we show that combinatorial, scale-space selective fingerprinting can successfully encode fin individuality. We then measure the species-specific distribution of visual individuality along the fin contour via an embedding into a global ‘fin space’. Exploiting this domain, we finally propose a non-linear model for individual animal recognition and combine all approaches into a fine-grained multi-instance framework. We provide a system evaluation, compare results to prior work, and report performance and properties in detail.

The objective of this paper is the fully automated visual identification of individual Holstein Friesian cattle from dorsal RGB-D imagery taken in real-world farm environments. Autonomous and non-intrusive cattle identification could provide an essential tool for economically-viable machinised farming analytics, social monitoring, cattle traceability, food production management and more. We contribute a dataset and propose a system that can reliably derive animal identities from top-down stills by first depth-segmenting animals in RGB-D frames, and then extracting a subset of local ASIFT coat descriptors predicted as sufficiently individually distinctive across the species. Predictions are generated by a support vector machine (SVM) using radial basis function (RBF) kernels for predictions based on the ASIFT descriptor structure. We show that learning such a species-specific ID-model is effective, and we demonstrate robustness to poor or complex input image conditions such as more than one cow present, bad depth segmentation, etc. The proposed system yields 97% identification accuracy over testing on approximately 86,000 image pair comparisons covering a herd of 40 individuals from the FriesianCattle2015 Dataset.

Quantitative assessment of the quality of motion is increasingly in demand by clinicians in healthcare and rehabilitation monitoring of patients. We study and compare the performances of different pose representations and HMM models of dynamics of movement for online quality assessment of human motion. In a general sense, our assessment framework builds a model of normal human motion from skeleton-based samples of healthy individuals. It encapsulates the dynamics of human body pose using robust manifold representation and a first-order Markovian assumption. We then assess deviations from it via a continuous online measure. We compare different feature representations, reduced dimensionality spaces, and HMM models on motions typically tested in clinical settings, such as gait on stairs and flat surfaces, and transitions between sitting and standing. Our dataset is manually labelled by a qualified physiotherapist. The continuous-state HMM, combined with pose representation based on body-joints’ location, outperforms standard discrete-state HMM approaches and other skeleton-based features in detecting gait abnormalities, as well as assessing deviations from the motion model on a frame-by-frame basis.

We present a real-time RGB-D object tracker which manages occlusions and scale changes in a wide variety of scenarios. Its accuracy matches, and in many cases outperforms, state-of-the-art algorithms for precision and it far exceeds most in speed. We build our algorithm on the existing colour-only KCF tracker which uses the `kernel trick' to extend correlation filters for fast tracking. We fuse colour and depth cues as the tracker's features, and furthermore, exploit the depth data to both adjust a given target's scale, and detect and manage occlusions in such a way as to maintain real-time performance, exceeding on average 40~fps. We benchmark our approach using 2 publicly available datasets and make our easy-to-extend modularised code available to other researchers.

The objective of this paper is automatically to identify individual great white sharks in a database of thousands of unconstrained fin images. The approach put forward appreciates shark fins in natural imagery as smooth, flexible and partially occluded objects with an individuality encoding trailing edge. In order to recover animal identities therefrom. We first introduce an open contour stroke model which extends multi-scale region segmentation to achieve robust fin detection. Secondly, we show that combinatorial spectral fingerprinting can successfully encode individuality in fin boundaries. We combine both approaches in a fine-grained multi-instance recognition framework. We provide a detailed evaluation of the system components and report its performance and properties.

Monitoring actions at home can provide essential information for rehabilitation management. This paper presents a comparative study and a dataset for the fully automated, sample-accurate recognition of common home actions in the living room environment using commercial-grade, inexpensive inertial and visual sensors. We investigate the practical home-use of body-worn mobile phone inertial sensors together with an Asus Xmotion RGB-Depth camera to achieve monitoring of daily living scenarios. To test this setup against realistic data, we introduce the challenging SPHERE-H130 action dataset containing 130 sequences of 13 household actions recorded in a home environment. We report automatic recognition results at maximal temporal resolution, which indicate that a vision-based approach outperforms accelerometer provided by two phone-based inertial sensors by an average of 14.85% accuracy for home actions. Further, we report improved accuracy of a vision-based approach over accelerometry on particularly challenging actions as well as when generalising across subjects.

We present an application that demonstrates a proof of concept system for automated in-the-field monitoring of disease in wheat crops. Such in-situ applications are required to be robust in the presence of clutter, provide rapid and accurate analysis and are able to operate at scale. We propose a processing pipeline that detects key wheat diseases in cluttered field imagery. First, we describe and evaluate a high dimensional texture descriptor combined with a randomised forest approach for automated primary leaf recognition. Second, we show that a combined nearest neighbour classifier and voting system applied to segmented leaf regions can robustly determine the presence and type of disease. The system has been tested on a real-world database of images of wheat leaves captured in-the-field using a standard smart phone.