Search

The human body represents a collection of interacting systems that range in scale from nanometers to meters. Investigations from a systems perspective focus on how the parts work together to enact changes across spatial scales, and further our understanding of how systems function and fail. Here, we highlight systems approaches presented at the 2022 Summer Biomechanics, Bio-engineering, and Biotransport Conference in the areas of solid mechanics; fluid mechanics; tissue and cellular engineering; biotransport; and design, dynamics, and rehabilitation; and biomechanics education. Systems approaches are yielding new insights into human biology by leveraging state-of-the-art tools, which could ultimately lead to more informed design of therapies and medical devices for preventing and treating disease as well as rehabilitating patients using strategies that are uniquely optimized for each patient. Educational approaches can also be designed to foster a foundation of systems-level thinking.

Current controls over diffusion models (e.g., through text or ControlNet) for image generation fall short in recognizing abstract, continuous attributes like illumination direction or non-rigid shape change. In this paper, we present an approach for allowing users of text-to-image models to have fine-grained control of several attributes in an image. We do this by engineering special sets of input tokens that can be transformed in a continuous manner -- we call them Continuous 3D Words. These attributes can, for example, be represented as sliders and applied jointly with text prompts for fine-grained control over image generation. Given only a single mesh and a rendering engine, we show that our approach can be adopted to provide continuous user control over several 3D-aware attributes, including time-of-day illumination, bird wing orientation, dollyzoom effect, and object poses. Our method is capable of conditioning image creation with multiple Continuous 3D Words and text descriptions simultaneously while adding no overhead to the generative process. Project Page: https://ttchengab.github.io/continuous_3d_words, Comment: Project Page: https://ttchengab.github.io/continuous_3d_words

El texto completo de este trabajo no se encuentra disponible por no haber sido facilitado aún por su autor, por restricciones de copyright, o por no existir una versión digital, Amphibian species are declining at an alarming rate on a global scale in large part owing to an infectious disease caused by the chytridiomycete fungus, Batrachochytrium dendrobatidis. This disease of amphibians has recently emerged within Europe, but knowledge of its effects on amphibian assemblages remains poor. Importantly, little is known about the environmental envelope that is associated with chytridiomycosis in Europe and the potential for climate change to drive future disease dynamics. Here, we use long-term observations on amphibian population dynamics in the Penalara Natural Park, Spain, to investigate the link between climate change and chytridiomycosis. Our analysis shows a significant association between change in local climatic variables and the occurrence of chytridiomycosis within this region. Specifically, we show that rising temperature is linked to the occurrence of chytrid-related disease, consistent with the chytrid-thermal-optimum hypothesis. We show that these local variables are driven by general circulation patterns, principally the North Atlantic Oscillation. Given that B. dendrobatidis is known to be broadly distributed across Europe, there is now an urgent need to assess the generality of our finding and determine whether climate-driven epidemics may be expected to impact on amphibian species across the wider region., Fundación BBVA, Natural Environment Research Council (NERC), Depto. de Física de la Tierra y Astrofísica, Fac. de Ciencias Físicas, TRUE, pub

The success of denoising diffusion models in representing rich data distributions over 2D raster images has prompted research on extending them to other data representations, such as vector graphics. Unfortunately due to their variable structure and scarcity of vector training data, directly applying diffusion models on this domain remains a challenging problem. Using workarounds like optimization via Score Distillation Sampling (SDS) is also fraught with difficulty, as vector representations are non trivial to directly optimize and tend to result in implausible geometries such as redundant or self-intersecting shapes. NIVeL addresses these challenges by reinterpreting the problem on an alternative, intermediate domain which preserves the desirable properties of vector graphics -- mainly sparsity of representation and resolution-independence. This alternative domain is based on neural implicit fields expressed in a set of decomposable, editable layers. Based on our experiments, NIVeL produces text-to-vector graphics results of significantly better quality than the state-of-the-art.

We present personalized residuals and localized attention-guided sampling for efficient concept-driven generation using text-to-image diffusion models. Our method first represents concepts by freezing the weights of a pretrained text-conditioned diffusion model and learning low-rank residuals for a small subset of the model's layers. The residual-based approach then directly enables application of our proposed sampling technique, which applies the learned residuals only in areas where the concept is localized via cross-attention and applies the original diffusion weights in all other regions. Localized sampling therefore combines the learned identity of the concept with the existing generative prior of the underlying diffusion model. We show that personalized residuals effectively capture the identity of a concept in ~3 minutes on a single GPU without the use of regularization images and with fewer parameters than previous models, and localized sampling allows using the original model as strong prior for large parts of the image., Comment: CVPR 2024. Project page at https://cusuh.github.io/personalized-residuals

Animated logos are a compelling and ubiquitous way individuals and brands represent themselves online. Manually authoring these logos can require significant artistic skill and effort. To help novice designers animate logos, design tools currently offer templates and animation presets. However, these solutions can be limited in their expressive range. Large language models have the potential to help novice designers create animated logos by generating animation code that is tailored to their content. In this paper, we introduce LogoMotion, an LLM-based system that takes in a layered document and generates animated logos through visually-grounded program synthesis. We introduce techniques to create an HTML representation of a canvas, identify primary and secondary elements, synthesize animation code, and visually debug animation errors. When compared with an industry standard tool, we find that LogoMotion produces animations that are more content-aware and are on par in terms of quality. We conclude with a discussion of the implications of LLM-generated animation for motion design.

Procedural noise is a fundamental component of computer graphics pipelines, offering a flexible way to generate textures that exhibit "natural" random variation. Many different types of noise exist, each produced by a separate algorithm. In this paper, we present a single generative model which can learn to generate multiple types of noise as well as blend between them. In addition, it is capable of producing spatially-varying noise blends despite not having access to such data for training. These features are enabled by training a denoising diffusion model using a novel combination of data augmentation and network conditioning techniques. Like procedural noise generators, the model's behavior is controllable via interpretable parameters and a source of randomness. We use our model to produce a variety of visually compelling noise textures. We also present an application of our model to improving inverse procedural material design; using our model in place of fixed-type noise nodes in a procedural material graph results in higher-fidelity material reconstructions without needing to know the type of noise in advance., Comment: In ACM Transactions on Graphics (Proceedings of SIGGRAPH) 2024, 21 pages

The brain's computational functions, from information synthesis and memory recall to consciousness, remain largely enigmatic. Despite great scientific effort, the mechanisms behind information storage, computation, and readout within the brain are still not well understood. Because of the warm, crowded nature of biological systems, it has traditionally been thought that quantum states would rapidly decohere and be unsuitable for biological information storage. Recently, the "Quantum Brain theory" has proposed the potential for a quantum information network in the brain that could maintain and process quantum information on timescales relevant to brain function. Within the framework of this theory, quantum information is stored in the collective phosphorus spin states of nanometric calcium phosphate clusters known as Posner molecules. The pair binding of these Posner molecules is postulated to be impacted by the collective phosphorus spin states, allowing for a unique quantum state to biochemical signal transduction mechanism. Given that this theory puts forward a uniquely possible method for quantum information processing in the brain, experimental validation or refutation of the theory would be an important step towards an understanding of brain function. This thesis presents experimental results aimed at testing the possibility of this Quantum Brain theory. We show that Posner molecules are stable in simulated body fluids, and uncover a differential lithium isotope effect on the growth of calcium phosphate. This isotope effect provides a link between nuclear spin states and calcium phosphate binding dynamics, and can be connected to in vivo lithium isotope effects in mitochondria. We also uncover spectroscopically 'dark state' phosphate assemblies after measuring unexpected 31P NMR signals, revealing that phosphate solution behavior and 31P nuclear spin dynamics are much more complex than previously predicted. These results verify key predictions of the Quantum Br

Phosphates play a ubiquitous role in biology, from structural components including cell membranes and bone to energy storage via ATP, but the solution phase space leading to the formation of phosphate clustering is not entirely understood. My work will help elucidate the role of phosphate clusters in biological solutions and the unresolved formation pathway of bone. Furthermore, my research will shed additional insight onto \textsuperscript{31}P nuclei quantum mechanical properties, specifically those found in Posner Clusters of molecular formula Ca\textsubscript{9}(PO\textsubscript{4})\textsubscript{6}, which are implicated to function as a platform for quantum computation in cognitive processes.This dissertation will show with solution nuclear magnetic resonance (NMR) experiments that simple phosphate species including orthophosphates, pyrophosphates, and adenosine phosphates associate into dynamic assemblies in dilute aqueous solutions that are spectroscopically "dark”, highlighting a hitherto unreported property of phosphate's native state in biological solutions. This dissertation will also discuss solid state NMR spin counting experiments of calcium phosphate species at varying time points in their structural evolution that were vitrified at 100 K and show calcium phosphate prenucleation clusters with a minimum of five dipolar coupled \textsuperscript{31}P atoms. These results provide a novel basis for the characterization of nonclassical growth pathways and led to developments in freeze quench instrumentation, also showcased here.

Using a citizen science approach, we identify a country-wide exposure to aerosolized spores of a human fungal pathogen, Aspergillus fumigatus, that has acquired resistance to the agricultural fungicide tebuconazole and first-line azole clinical antifungal drugs. Genomic analysis shows no distinction between resistant genotypes found in the environment and in patients, indicating that at least 40% of azole-resistant A. fumigatus infections are acquired from environmental exposures. Hotspots and coldspots of aerosolized azole-resistant spores were not stable between seasonal sampling periods. This suggests a high degree of atmospheric mixing resulting in an estimated per capita cumulative annual exposure of 21 days (±2.6). Because of the ubiquity of this measured exposure, it is imperative that we determine sources of azole-resistant A. fumigatus to reduce treatment failure in patients with aspergillosis.

AU Sexual: Pleaseconfirmthatallheadinglevelsarerepresentedcorrectly reproduction involving meiosis is essential in most eukaryotes. : This produces offspring with novel genotypes, both by segregation of parental chromosomes as well as crossovers between homologous chromosomes. A sexual cycle for the opportunistic human pathogenic fungus Aspergillus fumigatus is known, but the genetic consequences of meiosis have remained unknown. Among other Aspergilli, it is known that A. flavus has a moderately high recombination rate with an average of 4.2 crossovers per chromosome pair, whereas A. nidulans has in contrast a higher rate with 9.3 crossovers per chromosome pair. Here, we show in a cross between A. fumigatus strains that they produce an average of 29.9 crossovers per chromosome pair and large variation in total map length across additional strain crosses. This rate of crossovers per chromosome is more than twice that seen for any known organism, which we discuss in relation to other genetic model systems. We validate this high rate of crossovers through mapping of resistance to the laboratory antifungal acriflavine by using standing variation in an undescribed ABC efflux transporter. We then demonstrate that this rate of crossovers is sufficient to produce one of the common multidrug resistant haplotypes found in the cyp51A gene (TR34/L98H) in crosses among parents harboring either of 2 nearby genetic variants, possibly explaining the early spread of such haplotypes. Our results suggest that genomic studies in this species should reassess common assumptions about linkage between genetic regions. The finding of an unparalleled crossover rate in A. fumigatus provides opportunities to understand why these rates are not generally higher in other eukaryotes.

Representing a 3D shape with a set of primitives can aid perception of structure, improve robotic object manipulation, and enable editing, stylization, and compression of 3D shapes. Existing methods either use simple parametric primitives or learn a generative shape space of parts. Both have limitations: parametric primitives lead to coarse approximations, while learned parts offer too little control over the decomposition. We instead propose to decompose shapes using a library of 3D parts provided by the user, giving full control over the choice of parts. The library can contain parts with high-quality geometry that are suitable for a given category, resulting in meaningful decompositions with clean geometry. The type of decomposition can also be controlled through the choice of parts in the library. Our method works via a self-supervised approach that iteratively retrieves parts from the library and refines their placements. We show that this approach gives higher reconstruction accuracy and more desirable decompositions than existing approaches. Additionally, we show how the decomposition can be controlled through the part library by using different part libraries to reconstruct the same shapes., Comment: CVPR 2023

Genomic analyses are widely applied to epidemiological, population genetic and experimental studies of pathogenic fungi. A wide range of methods are employed to carry out these analyses, typically without including controls that gauge the accuracy of variant prediction. The importance of tracking outbreaks at a global scale has raised the urgency of establishing high-accuracy pipelines that generate consistent results between research groups. To evaluate currently employed methods for whole-genome variant detection and elaborate best practices for fungal pathogens, we compared how 14 independent variant calling pipelines performed across 35 Candida auris isolates from 4 distinct clades and evaluated the performance of variant calling, single-nucleotide polymorphism (SNP) counts and phylogenetic inference results. Although these pipelines used different variant callers and filtering criteria, we found high overall agreement of SNPs from each pipeline. This concordance correlated with site quality, as SNPs discovered by a few pipelines tended to show lower mapping quality scores and depth of coverage than those recovered by all pipelines. We observed that the major differences between pipelines were due to variation in read trimming strategies, SNP calling methods and parameters, and downstream filtration criteria. We calculated specificity and sensitivity for each pipeline by aligning three isolates with chromosomal level assemblies and found that the GATK-based pipelines were well balanced between these metrics. Selection of trimming methods had a greater impact on SAMtools-based pipelines than those using GATK. Phylogenetic trees inferred by each pipeline showed high consistency at the clade level, but there was more variability between isolates from a single outbreak, with pipelines that used more stringent cutoffs having lower resolution. This project generated two truth datasets useful for routine benchmarking of C. auris variant calling, a consensus VCF of g

Automatic generation of fonts can be an important aid to typeface design. Many current approaches regard glyphs as pixelated images, which present artifacts when scaling and inevitable quality losses after vectorization. On the other hand, existing vector font synthesis methods either fail to represent the shape concisely or require vector supervision during training. To push the quality of vector font synthesis to the next level, we propose a novel dual-part representation for vector glyphs, where each glyph is modeled as a collection of closed "positive" and "negative" path pairs. The glyph contour is then obtained by boolean operations on these paths. We first learn such a representation only from glyph images and devise a subsequent contour refinement step to align the contour with an image representation to further enhance details. Our method, named DualVector, outperforms state-of-the-art methods in vector font synthesis both quantitatively and qualitatively. Our synthesized vector fonts can be easily converted to common digital font formats like TrueType Font for practical use. The code is released at https://github.com/thuliu-yt16/dualvector., Comment: CVPR 2023

This report surveys advances in deep learning-based modeling techniques that address four different 3D indoor scene analysis tasks, as well as synthesis of 3D indoor scenes. We describe different kinds of representations for indoor scenes, various indoor scene datasets available for research in the aforementioned areas, and discuss notable works employing machine learning models for such scene modeling tasks based on these representations. Specifically, we focus on the analysis and synthesis of 3D indoor scenes. With respect to analysis, we focus on four basic scene understanding tasks -- 3D object detection, 3D scene segmentation, 3D scene reconstruction and 3D scene similarity. And for synthesis, we mainly discuss neural scene synthesis works, though also highlighting model-driven methods that allow for human-centric, progressive scene synthesis. We identify the challenges involved in modeling scenes for these tasks and the kind of machinery that needs to be developed to adapt to the data representation, and the task setting in general. For each of these tasks, we provide a comprehensive summary of the state-of-the-art works across different axes such as the choice of data representation, backbone, evaluation metric, input, output, etc., providing an organized review of the literature. Towards the end, we discuss some interesting research directions that have the potential to make a direct impact on the way users interact and engage with these virtual scene models, making them an integral part of the metaverse., Comment: Published in Computer Graphics Forum, Aug 2023

This project aimed to determine the recombination rate in additional Aspergillus fumigatus crosses by sequencing progeny of a sexual cross., This project aimed to determine the recombination rate in additional Aspergillus fumigatus crosses by sequencing progeny of a sexual cross.

How to frame (or crop) a photo often depends on the image subject and its context; e.g., a human portrait. Recent works have defined the subject-aware image cropping task as a nuanced and practical version of image cropping. We propose a weakly-supervised approach (GenCrop) to learn what makes a high-quality, subject-aware crop from professional stock images. Unlike supervised prior work, GenCrop requires no new manual annotations beyond the existing stock image collection. The key challenge in learning from this data, however, is that the images are already cropped and we do not know what regions were removed. Our insight is to combine a library of stock images with a modern, pre-trained text-to-image diffusion model. The stock image collection provides diversity and its images serve as pseudo-labels for a good crop, while the text-image diffusion model is used to out-paint (i.e., outward inpainting) realistic uncropped images. Using this procedure, we are able to automatically generate a large dataset of cropped-uncropped training pairs to train a cropping model. Despite being weakly-supervised, GenCrop is competitive with state-of-the-art supervised methods and significantly better than comparable weakly-supervised baselines on quantitative and qualitative evaluation metrics., Comment: AAAI 24. Extended version with supplemental materials

We present VecFusion, a new neural architecture that can generate vector fonts with varying topological structures and precise control point positions. Our approach is a cascaded diffusion model which consists of a raster diffusion model followed by a vector diffusion model. The raster model generates low-resolution, rasterized fonts with auxiliary control point information, capturing the global style and shape of the font, while the vector model synthesizes vector fonts conditioned on the low-resolution raster fonts from the first stage. To synthesize long and complex curves, our vector diffusion model uses a transformer architecture and a novel vector representation that enables the modeling of diverse vector geometry and the precise prediction of control points. Our experiments show that, in contrast to previous generative models for vector graphics, our new cascaded vector diffusion model generates higher quality vector fonts, with complex structures and diverse styles.

We introduce a novel image-to-painting method that facilitates the creation of large-scale, high-fidelity paintings with human-like quality and stylistic variation. To process large images and gain control over the painting process, we introduce a segmentation-based painting process and a dynamic attention map approach inspired by human painting strategies, allowing optimization of brush strokes to proceed in batches over different image regions, thereby capturing both large-scale structure and fine details, while also allowing stylistic control over detail. Our optimized batch processing and patch-based loss framework enable efficient handling of large canvases, ensuring our painted outputs are both aesthetically compelling and functionally superior as compared to previous methods, as confirmed by rigorous evaluations. Code available at: https://github.com/manuelladron/semantic\_based\_painting.git, Comment: 8 pages

Exploring variations of 3D shapes is a time-consuming process in traditional 3D modeling tools. Deep generative models of 3D shapes often feature continuous latent spaces that can, in principle, be used to explore potential variations starting from a set of input shapes. In practice, doing so can be problematic: latent spaces are high dimensional and hard to visualize, contain shapes that are not relevant to the input shapes, and linear paths through them often lead to sub-optimal shape transitions. Furthermore, one would ideally be able to explore variations in the original high-quality meshes used to train the generative model, not its lower-quality output geometry. In this paper, we present a method to explore variations among a given set of landmark shapes by constructing a mapping from an easily-navigable 2D exploration space to a subspace of a pre-trained generative model. We first describe how to find a mapping that spans the set of input landmark shapes and exhibits smooth variations between them. We then show how to turn the variations in this subspace into deformation fields, to transfer those variations to high-quality meshes for the landmark shapes. Our results show that our method can produce visually-pleasing and easily-navigable 2D exploration spaces for several different shape categories, especially as compared to prior work on learning deformation spaces for 3D shapes., Comment: SIGGRAPH Asia 2023, 15 pages

The fungal kingdom includes at least 6 million eukaryotic species and is remarkable with respect to its profound impact on global health, biodiversity, ecology, agriculture, manufacturing, and biomedical research. Approximately 625 fungal species have been reported to infect vertebrates, 200 of which can be human associated, either as commensals and members of our microbiome or as pathogens that cause infectious diseases. These organisms pose a growing threat to human health with the global increase in the incidence of invasive fungal infections, prevalence of fungal allergy, and the evolution of fungal pathogens resistant to some or all current classes of antifungals. More broadly, there has been an unprecedented and worldwide emergence of fungal pathogens affecting animal and plant biodiversity. Approximately 8,000 species of fungi and Oomycetes are associated with plant disease. Indeed, across agriculture, such fungal diseases of plants include new devastating epidemics of trees and jeopardize food security worldwide by causing epidemics in staple and commodity crops that feed billions. Further, ingestion of mycotoxins contributes to ill health and causes cancer. Coordinated international research efforts, enhanced technology translation, and greater policy outreach by scientists are needed to more fully understand the biology and drivers that underlie the emergence of fungal diseases and to mitigate against their impacts. Here, we focus on poignant examples of emerging fungal threats in each of three areas: human health, wildlife biodiversity, and food security.

Compost is an ecological niche for Aspergillus fumigatus due to its role as a decomposer of organic matter and its ability to survive the high temperatures associated with the composting process. Subsequently, composting facilities are associated with high levels of A. fumigatus spores that are aerosolized from compost and cause respiratory illness in workers. In the UK, gardening is an activity enjoyed by individuals of all ages, and it is likely that they are being exposed to A. fumigatus spores when handling commercial compost or compost they have produced themselves. In the present study, 246 citizen scientists collected 509 soil samples from locations in their gardens in the UK, from which were cultured 5,174 A. fumigatus isolates. Of these isolates, 736 (14%) were resistant to tebuconazole: the third most-sprayed triazole fungicide in the UK, which confers cross-resistance to the medical triazoles used to treat A. fumigatus lung infections in humans. These isolates were found to contain the common resistance mechanisms in the A. fumigatus cyp51A gene TR34/L98H or TR46/Y121F/T289A, as well as the less common resistance mechanisms TR34, TR53, TR46/Y121F/T289A/S363P/I364V/G448S, and (TR46)2/Y121F/M172I/T289A/G448S. Regression analyses found that soil samples containing compost were significantly more likely to grow tebuconazole-susceptible and tebuconazole-resistant A. fumigatus strains than those that did not and that compost samples grew significantly higher numbers of A. fumigatus than other samples.

This thesis is devoted to the study of quantum dynamics interspersed with quantum measurements. Focusing on a one-dimensional “hybrid” quantum circuit model consisting of random unitary gates and local projective measurements, we provide extensive evidence for a stable “weakly measured phase” that exhibits volume law entanglement entropy, and a novel continuous quantum dynamical phase transition between the weakly measured phase and a strongly measured “quantum Zeno phase” that exhibits area law entanglement entropy. We further study consequences of conformal symmetry at the critical point, as well as properties of the weakly measured phase. In the latter case, we develop an effective domain wall theory which correctly accounts for an unusual subvolume powerlaw correction to the entanglement entropy, and draw close connections between the domain wall theory and an emergent quantum error correcting code. In the last Chapter, we discuss an experimental protocol that can avoid the so- called “postselection problem” and allow scalable experimental observations of such transitions on near term quantum processors.

The fungal kingdom represents an extraordinary diversity of organisms with profound impacts across animal, plant, and ecosystem health. Fungi simultaneously support life, by forming beneficial symbioses with plants and producing life-saving medicines, and bring death, by causing devastating diseases in humans, plants, and animals. With climate change, increased antimicrobial resistance, global trade, environmental degradation, and novel viruses altering the impact of fungi on health and disease, developing new approaches is now more crucial than ever to combat the threats posed by fungi and to harness their extraordinary potential for applications in human health, food supply, and environmental remediation. To address this aim, the Canadian Institute for Advanced Research (CIFAR) and the Burroughs Wellcome Fund convened a workshop to unite leading experts on fungal biology from academia and industry to strategize innovative solutions to global challenges and fungal threats. This report provides recommendations to accelerate fungal research and highlights the major research advances and ideas discussed at the meeting pertaining to 5 major topics: (1) Connections between fungi and climate change and ways to avert climate catastrophe; (2) Fungal threats to humans and ways to mitigate them; (3) Fungal threats to agriculture and food security and approaches to ensure a robust global food supply; (4) Fungal threats to animals and approaches to avoid species collapse and extinction; and (5) Opportunities presented by the fungal kingdom, including novel medicines and enzymes.

Kākāpō are a critically endangered species of parrots restricted to a few islands off the coast of New Zealand. Kākāpō are very closely monitored, especially during nesting seasons. In 2019, during a highly successful nesting season, an outbreak of aspergillosis affected 21 individuals and led to the deaths of 9, leaving a population of only 211 kākāpō. In monitoring this outbreak, cultures of aspergillus were grown, and genome sequenced. These sequences demonstrate that, very unusually for an aspergillus outbreak, a single strain of aspergillus caused the outbreak. This strain was found on two islands, but only one had an outbreak of aspergillosis; indicating that the strain was necessary, but not sufficient, to cause disease. Our analysis provides an understanding of the 2019 outbreak and provides potential ways to manage such events in the future.

Mountains are an essential component of the global life-support system. They are characterized by a rugged, heterogenous landscape with rapidly changing environmental conditions providing myriad ecological niches over relatively small spatial scales. Although montane species are well adapted to life at extremes, they are highly vulnerable to human derived ecosystem threats. Here we build on the manifesto 'World Scientists' Warning to Humanity', issued by the Alliance of World Scientists, to outline the major threats to mountain ecosystems. We highlight climate change as the greatest threat to mountain ecosystems, which are more impacted than their lowland counterparts. We further discuss the cascade of "knock-on" effects of climate change such as increased UV radiation, altered hydrological cycles, and altered pollution profiles; highlighting the biological and socio-economic consequences. Finally, we present how intensified use of mountains leads to overexploitation and abstraction of water, driving changes in carbon stock, reducing biodiversity, and impacting ecosystem functioning. These perturbations can provide opportunities for invasive species, parasites and pathogens to colonize these fragile habitats, driving further changes and losses of micro- and macro-biodiversity, as well further impacting ecosystem services. Ultimately, imbalances in the normal functioning of mountain ecosystems will lead to changes in vital biological, biochemical, and chemical processes, critically reducing ecosystem health with widespread repercussions for animal and human wellbeing. Developing tools in species/habitat conservation and future restoration is therefore essential if we are to effectively mitigate against the declining health of mountains.

Invasive fungal infections pose an important threat to public health and are an under-recognized component of antimicrobial resistance, an emerging crisis worldwide. Across a period of profound global environmental change and expanding at-risk populations, human-infecting pathogenic fungi are evolving resistance to all licensed systemic antifungal drugs. In this Review, we highlight the main mechanisms of antifungal resistance and explore the similarities and differences between bacterial and fungal resistance to antimicrobial control. We discuss the research and innovation topics that are needed for risk reduction strategies aimed at minimizing the emergence of resistance in pathogenic fungi. These topics include links between the environment and One Health, surveillance, diagnostics, routes of transmission, novel therapeutics and methods to mitigate hotspots for fungal adaptation. We emphasize the global efforts required to steward our existing antifungal armamentarium, and to direct the research and development of future therapies and interventions. The impacts of fungal infections on human health are of increasing concern, and resistance of pathogenic fungi to all licensed systemic antifungals has been documented. In this Review, Fisher, Verweij and colleagues discuss the research and innovation topics that are needed to understand and minimize the occurrence and impact of antifungal resistance.

Diffusion models currently achieve state-of-the-art performance for both conditional and unconditional image generation. However, so far, image diffusion models do not support tasks required for 3D understanding, such as view-consistent 3D generation or single-view object reconstruction. In this paper, we present RenderDiffusion, the first diffusion model for 3D generation and inference, trained using only monocular 2D supervision. Central to our method is a novel image denoising architecture that generates and renders an intermediate three-dimensional representation of a scene in each denoising step. This enforces a strong inductive structure within the diffusion process, providing a 3D consistent representation while only requiring 2D supervision. The resulting 3D representation can be rendered from any view. We evaluate RenderDiffusion on FFHQ, AFHQ, ShapeNet and CLEVR datasets, showing competitive performance for generation of 3D scenes and inference of 3D scenes from 2D images. Additionally, our diffusion-based approach allows us to use 2D inpainting to edit 3D scenes., Comment: Accepted at CVPR 2023. Project page: https://github.com/Anciukevicius/RenderDiffusion

We introduce the task of spotting temporally precise, fine-grained events in video (detecting the precise moment in time events occur). Precise spotting requires models to reason globally about the full-time scale of actions and locally to identify subtle frame-to-frame appearance and motion differences that identify events during these actions. Surprisingly, we find that top performing solutions to prior video understanding tasks such as action detection and segmentation do not simultaneously meet both requirements. In response, we propose E2E-Spot, a compact, end-to-end model that performs well on the precise spotting task and can be trained quickly on a single GPU. We demonstrate that E2E-Spot significantly outperforms recent baselines adapted from the video action detection, segmentation, and spotting literature to the precise spotting task. Finally, we contribute new annotations and splits to several fine-grained sports action datasets to make these datasets suitable for future work on precise spotting., Comment: ECCV 2022; Website URL: https://jhong93.github.io/projects/spot.html

We present ASSET, a neural architecture for automatically modifying an input high-resolution image according to a user's edits on its semantic segmentation map. Our architecture is based on a transformer with a novel attention mechanism. Our key idea is to sparsify the transformer's attention matrix at high resolutions, guided by dense attention extracted at lower image resolutions. While previous attention mechanisms are computationally too expensive for handling high-resolution images or are overly constrained within specific image regions hampering long-range interactions, our novel attention mechanism is both computationally efficient and effective. Our sparsified attention mechanism is able to capture long-range interactions and context, leading to synthesizing interesting phenomena in scenes, such as reflections of landscapes onto water or flora consistent with the rest of the landscape, that were not possible to generate reliably with previous convnets and transformer approaches. We present qualitative and quantitative results, along with user studies, demonstrating the effectiveness of our method., Comment: SIGGRAPH 2022 - Journal Track

Approaches for single-view reconstruction typically rely on viewpoint annotations, silhouettes, the absence of background, multiple views of the same instance, a template shape, or symmetry. We avoid all such supervision and assumptions by explicitly leveraging the consistency between images of different object instances. As a result, our method can learn from large collections of unlabelled images depicting the same object category. Our main contributions are two ways for leveraging cross-instance consistency: (i) progressive conditioning, a training strategy to gradually specialize the model from category to instances in a curriculum learning fashion; and (ii) neighbor reconstruction, a loss enforcing consistency between instances having similar shape or texture. Also critical to the success of our method are: our structured autoencoding architecture decomposing an image into explicit shape, texture, pose, and background; an adapted formulation of differential rendering; and a new optimization scheme alternating between 3D and pose learning. We compare our approach, UNICORN, both on the diverse synthetic ShapeNet dataset - the classical benchmark for methods requiring multiple views as supervision - and on standard real-image benchmarks (Pascal3D+ Car, CUB) for which most methods require known templates and silhouette annotations. We also showcase applicability to more challenging real-world collections (CompCars, LSUN), where silhouettes are not available and images are not cropped around the object., Comment: ECCV 2022. Project webpage with code and videos: http://imagine.enpc.fr/~monniert/UNICORN

Rigged puppets are one of the most prevalent representations to create 2D character animations. Creating these puppets requires partitioning characters into independently moving parts. In this work, we present a method to automatically identify such articulated parts from a small set of character poses shown in a sprite sheet, which is an illustration of the character that artists often draw before puppet creation. Our method is trained to infer articulated parts, e.g. head, torso and limbs, that can be re-assembled to best reconstruct the given poses. Our results demonstrate significantly better performance than alternatives qualitatively and quantitatively.Our project page https://zhan-xu.github.io/parts/ includes our code and data.

[EN]Microplastics (MPs), a new class of pollutants that pose a threat to aquatic biodiversity, are of increasing global concern. In tandem, the amphibian chytrid fungus Batrachochytrium dendrobatidis (Bd) causing the disease chytridiomycosis is emerging worldwide as a major stressor to amphibians. We here assess whether synergies exist between this infectious disease and MP pollution by mimicking natural contact of a highly susceptible species (midwife toads, Alytes obstetricans) with a Bd-infected reservoir species (fire salamanders, Salamandra salamandra) in the presence and absence of MPs. We found that MP ingestion increases the burden of infection by Bd in a dose-dependent manner. However, MPs accumulated to a greater extent in amphibians that were not exposed to Bd, likely due to Bd-damaged tadpole mouthparts interfering with MP ingestion. Our experimental approach showed compelling interactions between two emergent processes, chytridiomycosis and MP pollution, necessitating further research into potential synergies between these biotic and abiotic threats to amphibians.

We explore a class of random tensor network models with "stabilizer" local tensors which we name Random Stabilizer Tensor Networks (RSTNs). For RSTNs defined on a two-dimensional square lattice, we perform extensive numerical studies of entanglement phase transitions between volume-law and area-law entangled phases of the one-dimensional boundary states. These transitions occur when either (a) the bond dimension $D$ of the constituent tensors is varied, or (b) the tensor network is subject to random breaking of bulk bonds, implemented by forced measurements. In the absence of broken bonds, we find that the RSTN supports a volume-law entangled boundary state with bond dimension $D\geq3$ where $D$ is a prime number, and an area-law entangled boundary state for $D=2$. Upon breaking bonds at random in the bulk with probability $p$, there exists a critical measurement rate $p_c$ for each $D\geq 3$ above which the boundary state becomes area-law entangled. To explore the conformal invariance at these entanglement transitions for different prime $D$, we consider tensor networks on a finite rectangular geometry with a variety of boundary conditions, and extract universal operator scaling dimensions via extensive numerical calculations of the entanglement entropy, mutual information and mutual negativity at their respective critical points. Our results at large $D$ approach known universal data of percolation conformal field theory, while showing clear discrepancies at smaller $D$, suggesting a distinct entanglement transition universality class for each prime $D$. We further study universal entanglement properties in the volume-law phase and demonstrate quantitative agreement with the recently proposed description in terms of a directed polymer in a random environment., Comment: Updated to published version

The ability to detect and monitor infectious disease in a phylogenetically informative manner is critical for their management. Phylogenetically informative diagnostic tests enable patterns of pathogen introduction or changes in the distribution of genotypes to be measured, enabling research into the ecology of the pathogen. Batrachochytrium dendrobatidis (Bd), a causative agent of chytridiomycosis in amphibian populations, emerged worldwide in the 21st century and is composed of six lineages which are display varying levels of virulence in their hosts. Research into the distribution, ecology and pathogenicity of these lineages has been hampered by an inability to type lineage efficiently. Here, we describe a lineage-specific TaqMan qPCR assay that differentiates the two lineages of Bd most commonly associated with chytridiomycosis: BdGPL and BdCAPE. We demonstrate how this assay can be used for the surveillance of wild populations of amphibians in Southern Africa using skin swabs, tissue samples and cultured isolates.

This paper introduces a model for producing stylized line drawings from 3D shapes. The model takes a 3D shape and a viewpoint as input, and outputs a drawing with textured strokes, with variations in stroke thickness, deformation, and color learned from an artist's style. The model is fully differentiable. We train its parameters from a single training drawing of another 3D shape. We show that, in contrast to previous image-based methods, the use of a geometric representation of 3D shape and 2D strokes allows the model to transfer important aspects of shape and texture style while preserving contours. Our method outputs the resulting drawing in a vector representation, enabling richer downstream analysis or editing in interactive applications., Comment: Accepted to ICCV 2021

Microplastics (MPs), a new class of pollutants that pose a threat to aquatic biodiversity, are of increasing global concern. In tandem, the amphibian chytrid fungus Batrachochytrium dendrobatidis (Bd) causing the disease chytridiomycosis is emerging worldwide as a major stressor to amphibians. We here assess whether synergies exist between this infectious disease and MP pollution by mimicking natural contact of a highly susceptible species (midwife toads, Alytes obstetricans) with a Bd-infected reservoir species (fire salamanders, Salamandra salamandra) in the presence and absence of MPs. We found that MP ingestion increases the burden of infection by Bd in a dose‑dependent manner. However, MPs accumulated to a greater extent in amphibians that were not exposed to Bd, likely due to Bd‑damaged tadpole mouthparts interfering with MP ingestion. Our experimental approach showed compelling interactions between two emergent processes, chytridiomycosis and MP pollution, necessitating further research into potential synergies between these biotic and abiotic threats to amphibians.

© 2020 IEEE. Many tasks in graphics and vision demand machinery for converting shapes into consistent representations with sparse sets of parameters; these representations facilitate rendering, editing, and storage. When the source data is noisy or ambiguous, however, artists and engineers often manually construct such representations, a tedious and potentially time-consuming process. While advances in deep learning have been successfully applied to noisy geometric data, the task of generating parametric shapes has so far been difficult for these methods. Hence, we propose a new framework for predicting parametric shape primitives using deep learning. We use distance fields to transition between shape parameters like control points and input data on a pixel grid. We demonstrate efficacy on 2D and 3D tasks, including font vectorization and surface abstraction.

Human pose is a useful feature for fine-grained sports action understanding. However, pose estimators are often unreliable when run on sports video due to domain shift and factors such as motion blur and occlusions. This leads to poor accuracy when downstream tasks, such as action recognition, depend on pose. End-to-end learning circumvents pose, but requires more labels to generalize. We introduce Video Pose Distillation (VPD), a weakly-supervised technique to learn features for new video domains, such as individual sports that challenge pose estimation. Under VPD, a student network learns to extract robust pose features from RGB frames in the sports video, such that, whenever pose is considered reliable, the features match the output of a pretrained teacher pose detector. Our strategy retains the best of both pose and end-to-end worlds, exploiting the rich visual patterns in raw video frames, while learning features that agree with the athletes' pose and motion in the target video domain to avoid over-fitting to patterns unrelated to athletes' motion. VPD features improve performance on few-shot, fine-grained action recognition, retrieval, and detection tasks in four real-world sports video datasets, without requiring additional ground-truth pose annotations., Comment: ICCV 2021 (poster)

Generating font glyphs of consistent style from one or a few reference glyphs, i.e., font completion, is an important task in topographical design. As the problem is more well-defined than general image style transfer tasks, thus it has received interest from both vision and machine learning communities. Existing approaches address this problem as a direct image-to-image translation task. In this work, we innovate to explore the generation of font glyphs as 2D graphic objects with the graph as an intermediate representation, so that more intrinsic graphic properties of font styles can be captured. Specifically, we formulate a cross-modality cycled image-to-image model structure with a graph constructor between an image encoder and an image renderer. The novel graph constructor maps a glyph's latent code to its graph representation that matches expert knowledge, which is trained to help the translation task. Our model generates improved results than both image-to-image baseline and previous state-of-the-art methods for glyph completion. Furthermore, the graph representation output by our model also provides an intuitive interface for users to do local editing and manipulation. Our proposed cross-modality cycled representation learning has the potential to be applied to other domains with prior knowledge from different data modalities. Our code is available at https://github.com/VITA-Group/Font_Completion_Graph., Comment: submitted to IEEE Transactions on Multimedia (TMM)

We investigate the problem of training generative models on a very sparse collection of 3D models. We use geometrically motivated energies to augment and thus boost a sparse collection of example (training) models. We analyze the Hessian of the as-rigid-as-possible (ARAP) energy to sample from and project to the underlying (local) shape space, and use the augmented dataset to train a variational autoencoder (VAE). We iterate the process of building latent spaces of VAE and augmenting the associated dataset, to progressively reveal a richer and more expressive generative space for creating geometrically and semantically valid samples. Our framework allows us to train generative 3D models even with a small set of good quality 3D models, which are typically hard to curate. We extensively evaluate our method against a set of strong baselines, provide ablation studies and demonstrate application towards establishing shape correspondences. We present multiple examples of interesting and meaningful shape variations even when starting from as few as 3-10 training shapes., Comment: CVPR 2022

Fonts are ubiquitous across documents and come in a variety of styles. They are either represented in a native vector format or rasterized to produce fixed resolution images. In the first case, the non-standard representation prevents benefiting from latest network architectures for neural representations; while, in the latter case, the rasterized representation, when encoded via networks, results in loss of data fidelity, as font-specific discontinuities like edges and corners are difficult to represent using neural networks. Based on the observation that complex fonts can be represented by a superposition of a set of simpler occupancy functions, we introduce \textit{multi-implicits} to represent fonts as a permutation-invariant set of learned implict functions, without losing features (e.g., edges and corners). However, while multi-implicits locally preserve font features, obtaining supervision in the form of ground truth multi-channel signals is a problem in itself. Instead, we propose how to train such a representation with only local supervision, while the proposed neural architecture directly finds globally consistent multi-implicits for font families. We extensively evaluate the proposed representation for various tasks including reconstruction, interpolation, and synthesis to demonstrate clear advantages with existing alternatives. Additionally, the representation naturally enables glyph completion, wherein a single characteristic font is used to synthesize a whole font family in the target style.