We propose a two-stage framework for motion in-betweening that combines diffusion-based motion generation with physics-based character adaptation. In Stage 1, a character-agnostic diffusion model synthesizes transitions from sparse keyframes on a canonical skeleton, allowing the same model to generalize across diverse characters. In Stage 2, a reinforcement learning-based controller adapts the canonical motion to the target character's morphology and dynamics, correcting artifacts and enhancing stylistic realism. This design supports scalable motion generation across characters with diverse skeletons without retraining the entire model. Experiments on standard benchmarks and stylized characters demonstrate that our method produces physically plausible, style-consistent motions under sparse and long-range constraints.
We present semisimp, a tool for semiautomatic simplification of three dimensional polygonal models. Existing automatic simplification technology is quite mature, but is not sensitive to the heightened importance of distinct semantic model regions such as faces and limbs, nor to simplification constraints imposed by model usage such as animation. semisimp allows users to preserve such regions by intervening in the simplification process. Users can manipulate the order in which basic simplifications are applied to redistribute model detail, improve the simplified models themselves by repositioning vertices with propagation to neighboring levels of detail, and adjust the hierarchical partitioning of the model surface to segment simplification and improve control of reordering and position propagation.
Takahiro Shirakawa, Tomoyuki Suzuki, Takuto Narumoto
et al.
We introduce MG-Gen, a framework that generates motion graphics directly from a single raster image. MG-Gen decompose a single raster image into layered structures represented as HTML, generate animation scripts for each layer, and then render them into a video. Experiments confirm MG-Gen generates dynamic motion graphics while preserving text readability and fidelity to the input conditions, whereas state-of-the-art image-to-video generation methods struggle with them. The code is available at https://github.com/CyberAgentAILab/MG-GEN.
Tanveer Khan Ibne Shafiq, Kamruzzaman Shaikh, Ferdous Rahman
et al.
Abstract Introduction The COVID-19 pandemic incurred numerous impediments on day-to-day emergency medical services including Opioid Substitution Therapy (OST) for People Who Inject Drugs (PWID). To prevent treatment cessation and lost to follow-up, we tried to implement an alternate mitigating intervention like telehealth. Methodology This research was conducted on a cohort of OST clients during the COVID-19 pandemic ( from 1st April 2020 to 31st March 2021) in Narayanganj, a port city adjacent to the capital Dhaka and one of the most COVID-affected districts, with a high PWID and HIV burden. The participants were male, female and transgender women who were all ex-PWID and were under OST services. A telehealth intervention model was designed and implemented in the OST clinic at Narayangonj. Quantitative data were collected during pre and post-intervention of telehealth services. Results A total of 297 OST clients of Narayangonj were provided with telehealth services from April 2020 to March 2021. The participants were predominantly male (98.7%), 37.7% were between 30–39 years of age. 39.4% of the telephone calls were related to COVID-19-related symptoms followed by 21.7% for opioid withdrawal, 12.5% for COVID-19 & vaccine-related information, 11.3% for chronic diseases like diabetes, hypertension and asthma, 9.3% for Skin and Soft Tissue Infection (SSTI), and 5.8% for methadone-related effects. There was an improvement in treatment retention (14.4% to 87%), loss to follow-up (20% to 8%), and overdose-related death (1.3% to 0%) from pre to post intervention of telehealth services. Conclusion From our experience, we found that the telehealth intervention is beneficial for the OST clients and thus ensures treatment continuity and retention, both of which serve as crucial success indicators of the OST programme. Using simply the mobile phone, this intervention can reduce structural and logistical needs like clinic spaces and fewer human resources, ensuring cost-effectiveness and value for money.
We derive methods to compute higher order differentials (Hessians and Hessian-vector products) of the rendering operator. Our approach is based on importance sampling of a convolution that represents the differentials of rendering parameters and shows to be applicable to both rasterization and path tracing. We further suggest an aggregate sampling strategy to importance-sample multiple dimensions of one convolution kernel simultaneously. We demonstrate that this information improves convergence when used in higher-order optimizers such as Newton or Conjugate Gradient relative to a gradient descent baseline in several inverse rendering tasks.
Transparency rendering is problematic and can be considered an open problem in real-time graphics. There are many different algorithms currently available, but handling complex scenes and achieving accurate, glitch-free results is still costly. This paper describes LucidRaster: a software rasterizer running on a GPU which allows for efficient exact rendering of complex transparent scenes. It uses a new two-stage sorting technique and sample accumulation method. On average it's faster than high-quality OIT approximations and only about 3x slower than hardware alpha blending. It can be very efficient especially when rendering scenes with high triangle density or high depth complexity.
Georges Gagneré, Andy Lavender, Cédric Plessiet
et al.
We describe1 two case studies of AvatarStaging theatrical mixed reality framework combining avatars and performers acting in an artistic context. We outline a qualitative approach toward the condition for stage presence for the avatars. We describe the motion control solutions we experimented with from the perspective of building a protocol of avatar direction in a mixed reality appropriate to live performance.
In this demo, we present a novel technique for approximating topologically optimal scaffoldings for 3D printed objects using a Monte Carlo algorithm based on the foraging behavior of the Physarum polycephalum slime mold. As a case study, we have created a biologically inspired bicycle helmet using this technique that is designed to be effective in resisting impacts. We have created a prototype of this helmet and propose further studies that measure the effectiveness and validity of the design.
Multi-sided surfaces are often defined by side interpolants (also called ribbons), i.e. the surface has to connect to the ribbons with a prescribed degree of smoothness. The I-patch is such a family of implicit surfaces capable of interpolating an arbitrary number of ribbons and can be used in design and approximation. While in the case of parametric surfaces describing ribbons is a well-discussed problem, defining implicit ribbons is a different task. This paper will introduce corner I-patches, a new representation that describes implicit surfaces based on corner interpolants. Those may be defined with much simpler surfaces, while the shape of the patch will depend on a handful of scalar parameters. Continuity between patches will be enforced via constraints on these parameters. Corner I-patches have several favorable properties that can be exploited for example in volume rendering or approximation.
Projection matrices are necessary for a large portion of rendering computer graphics. There are primarily two different types of projection matrices -- perspective and orthographic -- which are used frequently, and are traditionally treated as mutually incompatible with each other in how they are defined. Here, we bridge the gap between the two different forms of projection matrices to present a single generalized projection matrix that can represent both.
We present a novel representation of compressed data structure for simultaneous bounding volume hierarchy (BVH) traversals like they appear for instance in collision detection & proximity query. The main idea is to compress bounding volume (BV) descriptors and cluster BVH into a smaller parts 'treelet' that fit into CPU cache while at the same time maintain random-access and automatic cache-aware data structure layouts. To do that, we quantify BV and compress 'treelet' using predictor-corrector scheme with the predictor at a specific node in the BVH based on the chain of BVs upwards.
We present RodSteward, a design-to-assembly system for creating furniture-scale structures composed of 3D printed joints and precision-cut rods. The RodSteward systems consists of: RSDesigner, a fabrication-aware design interface that visualizes accurate geometries during edits and identifies infeasible designs; physical fabrication of parts via novel fully automatic construction of solid 3D-printable joint geometries and automatically generated cutting plans for rods; and RSAssembler, a guided-assembly interface that prompts the user to place parts in order while showing a focus+context visualization of the assembly in progress. We demonstrate the effectiveness of our tools with a number of example constructions of varying complexity, style and parameter choices.
In this article, we briefly describe various tools and approaches that algebraic geometry has to offer to straighten bent objects. Throughout this article we will consider a specific example of a bent or curved piece of paper which in our case acts very much like an elastica curve. We conclude this article with a suggestion to algebraic geometry as a viable and fast performance alternative of neural networks in vision and machine learning. The purpose of this article is not to build a full blown framework but to show possibility of using algebraic geometry as an alternative to neural networks for recognizing or extracting features on manifolds.
In this paper, we present RayTracer.jl, a renderer in Julia that is fully differentiable using source-to-source Automatic Differentiation (AD). This means that RayTracer not only renders 2D images from 3D scene parameters, but it can be used to optimize for model parameters that generate a target image in a Differentiable Programming (DP) pipeline. We interface our renderer with the deep learning library Flux for use in combination with neural networks. We demonstrate the use of this differentiable renderer in rendering tasks and in solving inverse graphics problems.
In this report we discuss and propose a correction to a convergence and stability issue occurring in the work of Da et al.[2015], in which they proposed a numerical model to simulate soap bubbles.
Sylvia L. Parmentier, Kristina Maier-Sam, Klaus Failing
et al.
AbstractThe apicomplexan parasite Sarcocystis calchasi (S. calchasi) triggers pigeon protozoal encephalitis, a neurologic disease in columbids. Accipiter hawks have been identified as the final host, and Columbidae and Psittaciformes as intermediate hosts. In this study, 368 free-ranging Accipiter hawks and 647 free-ranging common woodpigeons were sampled in a country-wide study in order to identify the prevalence of S. calchasi in these populations. A semi-nested PCR specific for S. calchasi tested positive in 7.3% (4.9–10.5) of submitted samples from Accipiter hawks. Juvenile Accipiter hawks (13.7%; 7.7–22.0) had a significantly higher infection rate with S. calchasi than adult Accipiter hawks (5.8%; 2.7–9.3). The prevalence of S. calchasi in common woodpigeons was 3.3% (5.4–9.7). Positive pigeons were identified in 14/16 federal states, and a region-dependency was detected, with higher rates of infection in the eastern parts of Germany. The results of this study suggest that the common woodpigeon is a natural reservoir for S. calchasi. In a study of one region for four consecutive years, an increase in prevalence was not detected. Findings indicate that the parasite is not newly introduced to Germany, but rather long established. The prevalence suggests that there is a substantial risk of S. calchasi infections in other free-ranging as well as captive host species.
We present a novel linear subdivision scheme for face-based tangent directional fields on triangle meshes. Our subdivision scheme is based on a novel coordinate-free representation of directional fields as halfedge-based scalar quantities, bridging the finite-element representation with discrete exterior calculus. By commuting with differential operators, our subdivision is structure-preserving: it reproduces curl-free fields precisely, and reproduces divergence-free fields in the weak sense. Moreover, our subdivision scheme directly extends to directional fields with several vectors per face by working on the branched covering space. Finally, we demonstrate how our scheme can be applied to directional-field design, advection, and robust earth mover's distance computation, for efficient and robust computation.
Approximation methods are widely used in many fields and many techniques have been published already. This comparative study presents a comparison of LOWESS (Locally weighted scatterplot smoothing) and RBF (Radial Basis Functions) approximation methods on noisy data as they use different approaches. The RBF approach is generally convenient for high dimensional scattered data sets. The LOWESS method needs finding a subset of nearest points if data are scattered. The experiments proved that LOWESS approximation gives slightly better results than RBF in the case of lower dimension, while in the higher dimensional case
How to establish the matching (or corresponding) between two different 3D shapes is a classical problem. This paper focused on the research on shape mapping of 3D mesh models, and proposed a shape mapping algorithm based on Hidden Markov Random Field and EM algorithm, as introducing a hidden state random variable associated with the adjacent blocks of shape matching when establishing HMRF. This algorithm provides a new theory and method to ensure the consistency of the edge data of adjacent blocks, and the experimental results show that the algorithm in this paper has a great improvement on the shape mapping of 3D mesh models.
Reseña del libro:Guerrero, Mauricio (ed.) (2014). Objetos públicos, espacios privados. Usuarios y relaciones sociales en tres centros comerciales de Santiago de Cali. Cali: Universidad Icesi, pp. 158.