Inexpensive and accessible sensors are accelerating data acquisition in animal ecology. These technologies hold great potential for large-scale ecological understanding, but are limited by current processing approaches which inefficiently distill data into relevant information. We argue that animal ecologists can capitalize on large datasets generated by modern sensors by combining machine learning approaches with domain knowledge. Incorporating machine learning into ecological workflows could improve inputs for ecological models and lead to integrated hybrid modeling tools. This approach will require close interdisciplinary collaboration to ensure the quality of novel approaches and train a new generation of data scientists in ecology and conservation.

Quantitative behavioral measurements are important for answering questions across scientific disciplines—from neuroscience to ecology. State-of-the-art deep-learning methods offer major advances in data quality and detail by allowing researchers to automatically estimate locations of an animal’s body parts directly from images or videos. However, currently available animal pose estimation methods have limitations in speed and robustness. Here, we introduce a new easy-to-use software toolkit, DeepPoseKit, that addresses these problems using an efficient multi-scale deep-learning model, called Stacked DenseNet, and a fast GPU-based peak-detection algorithm for estimating keypoint locations with subpixel precision. These advances improve processing speed >2x with no loss in accuracy compared to currently available methods. We demonstrate the versatility of our methods with multiple challenging animal pose estimation tasks in laboratory and field settings—including groups of interacting individuals. Our work reduces barriers to using advanced tools for measuring behavior and has broad applicability across the behavioral sciences.

Methods for collecting animal behaviour data in natural environments, such as direct observation and biologging, are typically limited in spatiotemporal resolution, the number of animals that can be observed and information about animals' social and physical environments. Video imagery can capture rich information about animals and their environments, but image‐based approaches are often impractical due to the challenges of processing large and complex multi‐image datasets and transforming resulting data, such as animals' locations, into geographical coordinates. We demonstrate a new system for studying behaviour in the wild that uses drone‐recorded videos and computer vision approaches to automatically track the location and body posture of free‐roaming animals in georeferenced coordinates with high spatiotemporal resolution embedded in contemporaneous 3D landscape models of the surrounding area. We provide two worked examples in which we apply this approach to videos of gelada monkeys and multiple species of group‐living African ungulates. We demonstrate how to track multiple animals simultaneously, classify individuals by species and age–sex class, estimate individuals' body postures (poses) and extract environmental features, including topography of the landscape and animal trails. By quantifying animal movement and posture while reconstructing a detailed 3D model of the landscape, our approach opens the door to studying the sensory ecology and decision‐making of animals within their natural physical and social environments.

Estimating animal populations is essential for conservation. Censusing large congregations is especially important since these are priorities for protection, but efficiently counting hundreds of thousands of moving animals remains a challenge. We developed a deep learning‐based system using consumer cameras that not only counts but also records behavioral information for large numbers of flying animals in a range of lighting conditions including near darkness. We built a robust training set without human labeling by leveraging data augmentation and background subtraction. We demonstrate this approach by estimating the size of a straw‐colored fruit bat (Eidolon helvum) colony in Kasanka National Park, Zambia with cameras encircling the colony to record evening emergence. Detection of bats was robust to deteriorating lighting conditions and changing backgrounds. Combined over five days, our population estimates ranged between 750,000 and 976,000 bats with a mean of 857,233. In addition to counts, we extracted wingbeat frequency, flight altitude, and local group polarity for 639,414 individuals. This open access method is an inexpensive but powerful approach that, in addition to radial emergences from central locations, can also be applied to unidirectional movements of flying groups, such as migratory streams of birds.

Quantitative behavioral measurements are important for answering questions across scientific disciplines—from neuroscience to ecology. State-of-the-art deep-learning methods offer major advances in data quality and detail by allowing researchers to automatically estimate locations of an animal’s body parts directly from images or videos. However, currently-available animal pose estimation methods have limitations in speed and robustness. Here we introduce a new easy-to-use software toolkit, DeepPoseKit , that addresses these problems using an eZcient multi-scale deep-learning model, called Stacked DenseNet , and a fast GPU-based peak-detection algorithm for estimating keypoint locations with subpixel precision. These advances improve processing speed >2× with no loss in accuracy compared to currently-available methods. We demonstrate the versatility of our methods with multiple challenging animal pose estimation tasks in laboratory and field settings—including groups of interacting individuals. Our work reduces barriers to using advanced tools for measuring behavior and has broad applicability across the behavioral sciences.