Search Results for author: Thomas W. Mitchel

Found 6 papers, 2 papers with code

Single Mesh Diffusion Models with Field Latents for Texture Generation

no code implementations14 Dec 2023 Thomas W. Mitchel, Carlos Esteves, Ameesh Makadia

We introduce a framework for intrinsic latent diffusion models operating directly on the surfaces of 3D shapes, with the goal of synthesizing high-quality textures.

Texture Synthesis

Möbius Convolutions for Spherical CNNs

1 code implementation28 Jan 2022 Thomas W. Mitchel, Noam Aigerman, Vladimir G. Kim, Michael Kazhdan

M\"obius transformations play an important role in both geometry and spherical image processing - they are the group of conformal automorphisms of 2D surfaces and the spherical equivalent of homographies.

Descriptive Image Segmentation +1

Field Convolutions for Surface CNNs

1 code implementation ICCV 2021 Thomas W. Mitchel, Vladimir G. Kim, Michael Kazhdan

We present a novel surface convolution operator acting on vector fields that is based on a simple observation: instead of combining neighboring features with respect to a single coordinate parameterization defined at a given point, we have every neighbor describe the position of the point within its own coordinate frame.


Efficient Spatially Adaptive Convolution and Correlation

no code implementations23 Jun 2020 Thomas W. Mitchel, Benedict Brown, David Koller, Tim Weyrich, Szymon Rusinkiewicz, Michael Kazhdan

Fast methods for convolution and correlation underlie a variety of applications in computer vision and graphics, including efficient filtering, analysis, and simulation.

Quotienting Impertinent Camera Kinematics for 3D Video Stabilization

no code implementations21 Mar 2019 Thomas W. Mitchel, Christian Wuelker, Jin Seob Kim, Sipu Ruan, Gregory S. Chirikjian

The foundation of this approach is a novel camera motion model that allows for real-world camera poses to be recovered directly from 3D motion fields.

Motion Estimation Video Stabilization

Signal Alignment for Humanoid Skeletons via the Globally Optimal Reparameterization Algorithm

no code implementations18 Jul 2018 Thomas W. Mitchel, Sipu Ruan, Gregory S. Chirikjian

Here, we introduce a variant of GORA for humanoid action recognition with skeleton sequences, which we call GORA-S. We briefly review the algorithm's mathematical foundations and contextualize them in the problem of action recognition with skeleton sequences.

Action Recognition Computational Efficiency +2

Cannot find the paper you are looking for? You can Submit a new open access paper.