Results 71 to 80 of about 14,765 (135)
SDFs from Unoriented Point Clouds using Neural Variational Heat Distances
We propose a novel variational approach for computing neural Signed Distance Fields (SDF) from unoriented point clouds. We first compute a small time step of heat flow (middle) and then use its gradient directions to solve for a neural SDF (right). Abstract We propose a novel variational approach for computing neural Signed Distance Fields (SDF) from ...
Samuel Weidemaier +5 more
wiley +1 more source
EvolvED: Evolutionary Embeddings to Understand the Generation Process of Diffusion Models
EvolvED visualises how diffusion models generate images by embedding intermediate outputs to preserve semantics and evolutionary structure. It supports analysis via (a) user‐defined goals and prompts, (b) sampling intermediate images, (c) extracting relevant features, and (d) visualising them in structured radial and rectilinear layouts for ...
Vidya Prasad +5 more
wiley +1 more source
A Survey of Methods for Constructing 3D Urban Models From Point Clouds
The survey outlines a general process for constructing 3D models from point clouds and categorizes the methods based on their use of templates, basic surface primitives, hybrid approaches, or linear primitives. Additionally, the survey reviews the datasets and benchmarks that are essential for testing and training the developed methods.
Chiara Romanengo +5 more
wiley +1 more source
Controllable Intrinsic Surface Pattern Generation Using Slime Mold Simulations
Abstract Surface‐based pattern simulations have proven valuable for texture design and scientific visualization, but existing methods face several limitations. Most simulations either target a narrow range of pattern types (e.g. spots, branching) or support a broad range of patterns at the cost of time‐consuming parameter tuning.
Jeffrey Layton +2 more
wiley +1 more source
Fast Injective Mesh Parameterization via Beltrami Coefficient Prolongation
Abstract We present a highly efficient and robust method for free boundary injective parameterization of disk‐like triangle meshes with low isometric distortion. Harmonic function–based approaches, grounded in a strong mathematical framework, are widely employed.
G. Fargion, O. Weber
wiley +1 more source
Self‐supervised Learning of Fine‐to‐Coarse Cuboid Shape Abstraction
Abstract The abstraction of 3D objects with simple geometric primitives like cuboids allows us to infer structural information from complex geometry. It is important for 3D shape understanding, structural analysis and geometric modeling. We introduce a novel fine‐to‐coarse self‐supervised learning approach to abstract collections of 3D shapes.
Gregor Kobsik +6 more
wiley +1 more source
Graph‐based Black and White Stylization
Abstract Stippling is an art style forming images from large numbers of small dots. Computer graphics practitioners have proposed numerous algorithms for stippling and stippling variants over the years. In this paper, we propose a black and white image stylization technique in which the image is formed from heterogeneous black and white vector marks ...
Ali Sattari Javid +2 more
wiley +1 more source
TreeON: Reconstructing 3D Tree Point Clouds from Orthophotos and Heightmaps
Abstract We present TreeON, a novel neural‐based framework for reconstructing detailed 3D tree point clouds from sparse top‐down geodata, using only a single orthophoto and its corresponding Digital Surface Model (DSM). Our method introduces a new training supervision strategy that combines both geometric supervision and a differentiable shadow and ...
Angeliki Grammatikaki +4 more
wiley +1 more source
LeafFit: Plant Assets Creation from 3D Gaussian Splatting
Abstract We propose LeafFit, a pipeline that converts 3D Gaussian Splatting (3DGS) of individual plants into editable, instanced mesh assets. While 3DGS faithfully captures complex foliage, its high memory footprint and lack of mesh topology make it incompatible with traditional game production workflows. We address this by leveraging the repetition of
Chang Luo, Nobuyuki Umetani
wiley +1 more source
Skeletal‐Driven Animation of Anatomical Humans via Neural Deformation Gradients
Abstract Most real‐time animation techniques for digital humans are limited to deforming the outer skin surface. Geometric skinning methods are highly efficient but struggle with artifacts such as collapsing joints or self‐intersections when animating inner anatomy along with the outer skin.
G. Nolte +3 more
wiley +1 more source

