Learning from multiple sensors is challenging due to spatio-temporal misalignment and differences in resolution and captured spectra. To that end, we introduce GeoWATCH, a flexible framework for training models on long sequences of satellite images sourced from multiple sensor platforms, which is designed to handle image classification, activity recognition, object detection, or object tracking tasks. Our system includes a novel partial weight loading mechanism based on sub-graph isomorphism which allows for continually training and modifying a network over many training cycles. This has allowed us to train a lineage of models over a long period of time, which we have observed has improved performance as we adjust configurations while maintaining a core backbone.

Tracking and classifying fish in optical underwater imagery presents several challenges which are encountered less frequently in terrestrial domains. Video may contain large schools comprised of many individuals, dynamic natural backgrounds, highly variable target scales, volatile collection conditions, and non-fish moving confusers including debris, marine snow, and other organisms. Additionally, there is a lack of large public datasets for algorithm evaluation available in this domain. The contributions of this paper is three fold. First, we present the FishTrack23 dataset which provides a large quantity of expert-annotated fish groundtruth tracks, in imagery and video collected across a range of different backgrounds, locations, collection conditions, and organizations. Approximately 850k bounding boxes across 26k tracks are included in the release of the ensemble, with potential for future growth in later releases. Second, we evaluate improvements upon baseline object detectors, trackers and classifiers on the dataset. Lastly, we integrate these methods into web and desktop interfaces to expedite annotation generation on new datasets.

Modern scientific, engineering and medical computational simulations, as well as experimental and observational data sensing/measuring devices, produce enormous amounts of data. While statistical analysis provides insight into this data, scientific visualization is tactically important for scientific discovery, product design and data analysis. These benefits are impeded, however, when scientific visualization algorithms are implemented from scratch — a time-consuming and redundant process in immersive application development. This process can greatly benefit from leveraging the state-of-the-art open-source Visualization Toolkit (VTK) and its community. Over the past two (almost three) decades, integrating VTK with a virtual reality (VR) environment has only been attempted to varying degrees of success. In this paper, we demonstrate two new approaches to simplify this amalgamation of an immersive interface with visualization rendering from VTK. In addition, we cover several enhancements to VTK that provide near real-time updates and efficient interaction. Finally, we demonstrate the combination of VTK with both Vrui and OpenVR immersive environments in example applications.

The availability of low-cost virtual reality (VR) systems coupled with a growing population of researchers accustomed to newer interface styles makes this a ripe time to help domain science researchers cross the bridge to utilizing immersive interfaces. The logical next step is for scientists, engineers, doctors, etc. to incorporate immersive visualization into their exploration and analysis workflows. However, from past experience, we know having access to equipment is not sufficient. There are also several software hurdles to overcome. Obstacles must be lowered to provide scientists, engineers, and medical professionals low-risk means of exploring technologies beyond their desktops.