Computer Vision for Autonomous UAV Flight Safety: An Overview and a Vision-based Safe Landing Pipeline Example

Recent years have seen an unprecedented spread of Unmanned Aerial Vehicles (UAVs, or “drones”), which are highly useful for both civilian and military applications. Flight safety is a crucial issue in UAV navigation, having to ensure accurate compliance with recently legislated rules and regulations. The emerging use of autonomous drones and UAV swarms raises additional issues, making it necessary to transfuse safety- and regulations-awareness to relevant algorithms and architectures. Computer vision plays a pivotal role in such autonomous functionalities. Although the main aspects of autonomous UAV technologies (e.g., path planning, navigation control, landing control, mapping and localization, target detection/tracking) are already mature and well-covered, ensuring safe flying in the vicinity of crowds, avoidance of passing over persons, or guaranteed emergency landing capabilities in case of malfunctions, are generally treated as an afterthought when designing autonomous UAV platforms for unstructured environments. This fact is reflected in the fragmentary coverage of the above issues in current literature. This overview attempts to remedy this situation, from the point of view of computer vision. It examines the field from multiple aspects, including regulations across the world and relevant current technologies. Finally, since very few attempts have been made so far towards a complete UAV safety flight and landing pipeline, an example computer vision-based UAV flight safety pipeline is introduced, taking into account all issues present in current autonomous drones. The content is relevant to any kind of autonomous drone flight (e.g., for movie/TV production, news-gathering, search and rescue, surveillance, inspection, mapping, wildlife monitoring, crowd monitoring/management), making this a topic of broad interest.

Betting on Bets

The AI Bookie column documents highlights from AI Bets, an online forum for the creation of adjudicatable predictions about the future of AI. Since the column’s inception 3 years ago, only a few scientific bets have been collected, despite universal approval around the idea of scientific betting. We hope to widen our reach with an additional first batch of seed bets that are of broad interest to the research community including AI bias, fifth sentence prediction, emotion regu-lation, big models, and fake news. For detailed guidelines and to place bets, visit sciencebets.org.

OptiDiff: structural variation detection from single optical mapping reads

Detecting structural variation (SV) in eukaryotic genomes is of broad interest due to its often dramatic phenotypic effects, but remains a major, costly challenge based on DNA sequencing data. A cost-effective alternative in detecting large-scale SV has become available with advances in optical mapping technology. However, the algorithmic approaches to identifying SVs from optical mapping data are limited. Here, we propose a novel, open-source SV detection tool, OptiDiff, which employs a single molecule based approach to detect and classify homozygous and heterozygous SVs at coverages as low as 20x, showing better performance than the state of the art.

Two-step nonlinear optical switch in a hydrogen-bonded perovskite-type crystal

Switchable nonlinear optical (NLO) materials have aroused broad interest on account of their captivating optical and electronic properties. We demonstrate a novel perovskite-type crystal with exceptional hydrogen bond interactions that...

Winning the NIST Contest: A scalable and general approach to differentially private synthetic data

We propose a general approach for differentially private synthetic data generation, that consists of three steps: (1) select a collection of low-dimensional marginals, (2) measure those marginals with a noise addition mechanism, and (3) generate synthetic data that preserves the measured marginals well. Central to this approach is Private-PGM, a post-processing method that is used to estimate a high-dimensional data distribution from noisy measurements of its marginals. We present two mechanisms, NIST-MST and MST, that are instances of this general approach. NIST-MST was the winning mechanism in the 2018 NIST differential privacy synthetic data competition, and MST is a new mechanism that can work in more general settings, while still performing comparably to NIST-MST. We believe our general approach should be of broad interest, and can be adopted in future mechanisms for synthetic data generation.

Early Mesoamerican Cities

Urbanization is a phenomenon that brings into focus a range of topics of broad interest to scholars. It is one of the central, enduring interests of anthropological archaeology. Because urbanization is a transformational process, it changes the relationships between social and cultural variables such as demography, economy, politics, and ideology. As one of a handful of cases in the ancient world where cities developed independently, Mesoamerica should play a major role in the global, comparative analysis of first-generation cities and urbanism in general. Yet most research focuses on later manifestations of urbanism in Mesoamerica, thereby perpetuating the fallacy that Mesoamerican cities developed relatively late in comparison to urban centers in the rest of the world. This volume presents new data, case studies, and models for approaching the subject of early Mesoamerican cities. It demonstrates how the study of urbanism in Mesoamerica, and all ancient civilizations, is entering a new and dynamic phase of scholarship.

HARLEY: A Semi-Automated detection of foci in fluorescence images of yeast

Quantification of cellular structures in fluorescence microscopy data is a key means of understanding cellular function. Unfortunately, numerous cellular structures present unique challenges in their ability to be unbiasedly and accurately detected and quantified. In our studies on stress granules in yeast, users displayed a striking variation of up to 3.7-fold in foci calls and were only able to replicate their results with 62-78% correlation, when requantifying the same images. To facilitate consistent results we developed HARLEY (Human Augmented Recognition of LLPS Ensembles in Yeast), a customizable software for detection and quantification of stress granules in S.cerevisiae. After a brief model training on ~20 cells the detection of foci is fully automated and based on closed loops in intensity contours, constrained only by the a-priori known size of the features of interest. Since no shape is implied, this method is not limited to round features, as is often the case with other algorithms. Candidate features are annotated with a set of geometrical and intensity-based properties to train a kernel Support Vector Machine to recognize features of interest. The trained classifier is then used to create consistent results across datasets. HARLEY is aimed at users without technical expertise, allows for batch processing and is freely available, which should be of broad interest to users focused on analysis of microscopy data in yeast.

Active math and grammar learning engages overlapping brain networks

We here demonstrate common neurocognitive long-term memory effects of active learning that generalize over course subjects (mathematics and vocabulary) by the use of fMRI. One week after active learning, relative to more passive learning, performance and fronto-parietal brain activity was significantly higher during retesting, possibly related to the formation and reactivation of semantic representations. These observations indicate that active learning conditions stimulate common processes that become part of the representations and can be reactivated during retrieval to support performance. Our findings are of broad interest and educational significance related to the emerging consensus of active learning as critical in promoting good long-term retention.

Why do we pick similar mates, or do we?

Humans often mate with those resembling themselves, a phenomenon described as positive assortative mating (PAM). The causes of this attract broad interest, but there is little agreement on the topic. This may be because empirical studies and reviews sometimes focus on just a few explanations, often based on disciplinary conventions. This review presents an interdisciplinary conceptual framework on the causes of PAM in humans, drawing on human and non-human biology, the social sciences, and the humanities. Viewing causality holistically, we first discuss the proximate causes (i.e. the ‘how’) of PAM, considering three mechanisms: stratification, convergence and mate choice. We also outline methods to control for confounders when studying mate choice. We then discuss ultimate explanations (i.e. ‘the why’) for PAM, including adaptive and non-adaptive processes. We conclude by suggesting a focus on interdisciplinarity in future research.

The Impact of Ambassador Animal Facilitated Programs on Visitor Curiosity and Connections: A Mixed-Methods Study

Many zoos and aquariums offer opportunities for visitors to have up-close encounters with ambassador animals; however, the impacts of these experiences on visitors’ connections to animals are not well documented. We used observations and family interviews in a sequential mixed-methods research study to explore how animal ambassador programs impact participants. We found that the type of ambassador animal did not affect the number of questions or comments made by participants during programs, suggesting broad interest in animals. Programs in which facilitators prompted participants with questions were especially successful at eliciting questions and guiding the topics of those questions, fueling deeper curiosity. Interviewees described themselves as “animal people” and self-reported practicing conservation behaviors, suggesting that their animal affiliative and pro-environmental identities could be leveraged to discuss conservation issues and encourage solutions-based behaviors. Lastly, opportunities to meet ambassador animals increased participants’ feelings of connecting to animals, building on previous research and corroborating findings. Programs should consider how to further build on these positive learning and affective outcomes by capitalizing on opportunities to provide deep insights into conservation issues and actions related to the ambassador animals.