Search Results for author:
Found 42 papers, 22 papers with code
MultiDiffusion: Fusing Diffusion Paths for Controlled Image Generation
In this work, we present MultiDiffusion, a unified framework that enables versatile and controllable image generation, using a pre-trained text-to-image diffusion model, without any further training or finetuning.
Volume Rendering of Neural Implicit Surfaces
Accurate sampling is important to provide a precise coupling of geometry and radiance; and (iii) it allows efficient unsupervised disentanglement of shape and appearance in volume rendering.
Multiview Neural Surface Reconstruction by Disentangling Geometry and Appearance
In this work we address the challenging problem of multiview 3D surface reconstruction.
Implicit Geometric Regularization for Learning Shapes
Representing shapes as level sets of neural networks has been recently proved to be useful for different shape analysis and reconstruction tasks.
Flow Matching for Generative Modeling
These paths are more efficient than diffusion paths, provide faster training and sampling, and result in better generalization.
Ranked #5 on
Density Estimation
on CIFAR-10
Flow Matching on General Geometries
To extend to general geometries, we rely on the use of spectral decompositions to efficiently compute premetrics on the fly.
Controlling Neural Level Sets
In turn, the sample network can be used to incorporate the level set samples into a loss function of interest.
SAL: Sign Agnostic Learning of Shapes from Raw Data
Recently, neural networks have been used as implicit representations for surface reconstruction, modelling, learning, and generation.
SALD: Sign Agnostic Learning with Derivatives
Learning 3D geometry directly from raw data, such as point clouds, triangle soups, or unoriented meshes is still a challenging task that feeds many downstream computer vision and graphics applications.
Riemannian Convex Potential Maps
Modeling distributions on Riemannian manifolds is a crucial component in understanding non-Euclidean data that arises, e. g., in physics and geology.
Point Convolutional Neural Networks by Extension Operators
This paper presents Point Convolutional Neural Networks (PCNN): a novel framework for applying convolutional neural networks to point clouds.
Ranked #80 on
3D Point Cloud Classification
on ModelNet40
Multi-chart Generative Surface Modeling
The new tensor data representation is used as input to Generative Adversarial Networks for the task of 3D shape generation.
Provably Powerful Graph Networks
It was shown that the popular message passing GNN cannot distinguish between graphs that are indistinguishable by the 1-WL test (Morris et al. 2018; Xu et al. 2019).
Ranked #6 on
Graph Classification
on COLLAB
Neural Conservation Laws: A Divergence-Free Perspective
We investigate the parameterization of deep neural networks that by design satisfy the continuity equation, a fundamental conservation law.
Set2Graph: Learning Graphs From Sets
Many problems in machine learning can be cast as learning functions from sets to graphs, or more generally to hypergraphs; in short, Set2Graph functions.
Generalized Schrödinger Bridge Matching
Modern distribution matching algorithms for training diffusion or flow models directly prescribe the time evolution of the marginal distributions between two boundary distributions.
Moser Flow: Divergence-based Generative Modeling on Manifolds
MF also produces a CNF via a solution to the change-of-variable formula, however differently from other CNF methods, its model (learned) density is parameterized as the source (prior) density minus the divergence of a neural network (NN).
Global Attention Improves Graph Networks Generalization
This paper advocates incorporating a Low-Rank Global Attention (LRGA) module, a computation and memory efficient variant of the dot-product attention (Vaswani et al., 2017), to Graph Neural Networks (GNNs) for improving their generalization power.
Ranked #1 on
Link Property Prediction
on ogbl-ddi
On Universal Equivariant Set Networks
The key theoretical tool used to prove the above results is an explicit characterization of all permutation equivariant polynomial layers.
Surface Networks via General Covers
Specifically, for the use case of learning spherical signals, our representation provides a low distortion alternative to several popular spherical parameterizations used in deep learning.
Phase Transitions, Distance Functions, and Implicit Neural Representations
Representing surfaces as zero level sets of neural networks recently emerged as a powerful modeling paradigm, named Implicit Neural Representations (INRs), serving numerous downstream applications in geometric deep learning and 3D vision.
Secondary Vertex Finding in Jets with Neural Networks
Jet classification is an important ingredient in measurements and searches for new physics at particle coliders, and secondary vertex reconstruction is a key intermediate step in building powerful jet classifiers.
High Energy Physics - Experiment High Energy Physics - Phenomenology
Photometric Stereo by Hemispherical Metric Embedding
We present a novel embedding method that maps pixels to normals on the unit hemisphere.
Wide baseline stereo matching with convex bounded-distortion constraints
This paper focuses on the challenging problem of finding correspondences once approximate epipolar constraints are given.
Invariant and Equivariant Graph Networks
In this paper we provide a characterization of all permutation invariant and equivariant linear layers for (hyper-)graph data, and show that their dimension, in case of edge-value graph data, is 2 and 15, respectively.
On the Universality of Invariant Networks
We conclude the paper by proving a necessary condition for the universality of $G$-invariant networks that incorporate only first-order tensors.
Wide Baseline Stereo Matching With Convex Bounded Distortion Constraints
This paper focuses on the challenging problem of finding correspondences once approximate epipolar constraints are given.
Isometric Autoencoders
Two sources of bad generalization are: extrinsic, where the learned manifold possesses extraneous parts that are far from the data; and intrinsic, where the encoder and decoder introduce arbitrary distortion in the low dimensional parameterization.
Augmenting Implicit Neural Shape Representations with Explicit Deformation Fields
Implicit neural representation is a recent approach to learn shape collections as zero level-sets of neural networks, where each shape is represented by a latent code.
Frame Averaging for Invariant and Equivariant Network Design
For example, Euclidean motion invariant/equivariant graph or point cloud neural networks.
Frame Averaging for Equivariant Shape Space Learning
A natural way to incorporate symmetries in shape space learning is to ask that the mapping to the shape space (encoder) and mapping from the shape space (decoder) are equivariant to the relevant symmetries.
Weisfeiler and Leman go Machine Learning: The Story so far
In recent years, algorithms and neural architectures based on the Weisfeiler--Leman algorithm, a well-known heuristic for the graph isomorphism problem, have emerged as a powerful tool for machine learning with graphs and relational data.
Matching Normalizing Flows and Probability Paths on Manifolds
Continuous Normalizing Flows (CNFs) are a class of generative models that transform a prior distribution to a model distribution by solving an ordinary differential equation (ODE).
Equivariant Polynomials for Graph Neural Networks
This paper introduces an alternative expressive power hierarchy based on the ability of GNNs to calculate equivariant polynomials of a certain degree.
VisCo Grids: Surface Reconstruction with Viscosity and Coarea Grids
Surface reconstruction has been seeing a lot of progress lately by utilizing Implicit Neural Representations (INRs).
Multisample Flow Matching: Straightening Flows with Minibatch Couplings
Simulation-free methods for training continuous-time generative models construct probability paths that go between noise distributions and individual data samples.
On Kinetic Optimal Probability Paths for Generative Models
We investigate Kinetic Optimal (KO) Gaussian paths and offer the following observations: (i) We show the KE takes a simplified form on the space of Gaussian paths, where the data is incorporated only through a single, one dimensional scalar function, called the \emph{data separation function}.
Bespoke Solvers for Generative Flow Models
For example, a Bespoke solver for a CIFAR10 model produces samples with Fr\'echet Inception Distance (FID) of 2. 73 with 10 NFE, and gets to 1% of the Ground Truth (GT) FID (2. 59) for this model with only 20 NFE.
Guided Flows for Generative Modeling and Decision Making
Classifier-free guidance is a key component for enhancing the performance of conditional generative models across diverse tasks.
Mosaic-SDF for 3D Generative Models
Current diffusion or flow-based generative models for 3D shapes divide to two: distilling pre-trained 2D image diffusion models, and training directly on 3D shapes.
D-Flow: Differentiating through Flows for Controlled Generation
Taming the generation outcome of state of the art Diffusion and Flow-Matching (FM) models without having to re-train a task-specific model unlocks a powerful tool for solving inverse problems, conditional generation, and controlled generation in general.
Bespoke Non-Stationary Solvers for Fast Sampling of Diffusion and Flow Models
This paper introduces Bespoke Non-Stationary (BNS) Solvers, a solver distillation approach to improve sample efficiency of Diffusion and Flow models.
