A5000795170- Advanced Numerical Analysis Techniques
- 3D Shape Modeling and Analysis
- Computer Graphics and Visualization Techniques
- Seismology and Earthquake Studies
- Earthquake Detection and Analysis
- Medical Image Segmentation Techniques
- earthquake and tectonic studies
- Advanced Vision and Imaging
- Business Process Modeling and Analysis
- Data Quality and Management
- Data Visualization and Analytics
- Neural Networks and Applications
- Image and Signal Denoising Methods
- Medical Imaging and Analysis
- Semantic Web and Ontologies
- Image Processing and 3D Reconstruction
- Video Analysis and Summarization
- Data Analysis with R
- Anomaly Detection Techniques and Applications
- Image Retrieval and Classification Techniques
- Advanced Data Compression Techniques
- Machine Learning and Data Classification
- Advanced Image Fusion Techniques
- Color Science and Applications
- Brain Tumor Detection and Classification
University of Coimbra
2022-2025
Institute for Systems Engineering and Computers
2025
Instituto Nacional de Matemática Pura e Aplicada
2023
Pontifical Catholic University of Rio de Janeiro
2018-2021
This work investigates the use of smooth neural networks for modeling dynamic variations implicit surfaces under level set equation (LSE). For this, it extends representation to space-time ℝ <sup xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">3</sup> × ℝ, which opens up mechanisms continuous geometric transformations. Examples include evolving an initial surface towards general vector fields, smoothing and sharpening using mean curvature equation,...
We present MR-Net, a general architecture for multiresolution neural networks, and framework imaging applications based on this architecture. Our coordinate-based networks are continuous both in space scale as they composed of multiple stages that progressively add finer details. Besides that, compact efficient representation. show examples image representation to texture magnification, minification, antialiasing.
Summary The objective of this work is to develop and train feedforward artificial neural networks (ANNs) on the forecasting layer permeability in heterogeneous reservoirs. results are validated by comparing model outputs with curves computed from production logging data. Production logs used as targets model. A flow-profile interpretation method compute continuous free wellbore skin effects. In addition, segmentation techniques applied high-resolution ultrasonic image logs. These provide not...
This survey presents methods that use neural networks for implicit representations of 3D geometry — functions. We explore the different aspects functions shape modeling and synthesis. aim to provide a theoretical analysis reconstruction using deep introduce discussion between researchers interested in this research field.
Download This Paper Open PDF in Browser Add to My Library Share: Permalink Using these links will ensure access this page indefinitely Copy URL DOI
Earth structural heterogeneities have a remarkable role in the petroleum economy for both exploration and production projects. Automatic detection of detailed is challenging when considering modern machine learning techniques like deep neural networks. Typically, these can be an excellent tool assisted interpretation such heterogeneities, but it heavily depends on amount data to trained. We propose efficient cost-effective architecture detecting seismic using Convolutional Neural Networks...
Earth structural heterogeneities have a remarkable role in the petroleum economy for both exploration and production projects. Automatic detection of detailed is challenging when considering modern machine learning techniques like deep neural networks. Typically, these can be an excellent tool assisted interpretation such heterogeneities, but it heavily depends on amount data to trained. We propose efficient cost-effective architecture detecting seismic using Convolutional Neural Networks...
Neural fields have emerged as a promising framework for representing different types of signals. This tutorial focus on the existing literature and shares practical insights derived from hands-on experimentation with neural fields, specifically in approximating implicit functions surfaces. Our emphasis lies strategies leveraging differential geometry concepts to enhance training outcomes showcase applications within this domain.
We introduce a novel approach for rendering static and dynamic 3D neural signed distance functions (SDF) in real-time. rely on nested neighborhoods of zero-level sets SDFs, mappings between them. This framework supports animations achieves real-time performance without the use spatial data-structures. It consists three uncoupled algorithms representing steps. The multiscale sphere tracing focuses minimizing iteration time by using coarse approximations earlier iterations. normal mapping...
We introduce a neural implicit framework that exploits the differentiable properties of networks and discrete geometry point-sampled surfaces to approximate them as level sets functions. To train function, we propose loss functional approximates signed distance allows terms with high-order derivatives, such alignment between principal directions curvature, learn more geometric details. During training, consider non-uniform sampling strategy based on curvatures surface prioritize points This...
We introduce a neural implicit framework that bridges discrete differential geometry of triangle meshes and continuous surfaces. It exploits the differentiable properties networks to approximate them as level sets functions . To train function, we propose loss functional allows terms with high-order derivatives, such alignment between principal directions curvature, learn more geometric details. During training, consider non-uniform sampling strategy based on curvatures mesh access points...
Face morphing is one of the seminal problems in computer graphics, with numerous artistic and forensic applications. It notoriously challenging due to pose, lighting, gender, ethnicity variations. Generally, this task consists a warping for feature alignment blending seamless transition between warped images. We propose leverage coordinate-based neural networks represent such warpings blendings face During training, we exploit smoothness flexibility networks, by combining energy functionals...