activity
20212024
most citedNCP: Neural Correspondence Prior for Effective Unsupervised Shape Matching

7 citations · 10 across the 10 of their papers we have counts for

collaborators

10 papers

cs.CV2024

PoNQ: a Neural QEM-based Mesh Representation

Nissim Maruani, Maks Ovsjanikov, Pierre Alliez +1

Although polygon meshes have been a standard representation in geometry processing, their irregular and combinatorial nature hinders their suitability for learning-based applicatio…

cs.CV20242 cited

Shape Non-rigid Kinematics (SNK): A Zero-Shot Method for Non-Rigid Shape Matching via Unsupervised Functional Map Regularized Reconstruction

Souhaib Attaiki, Maks Ovsjanikov

We present Shape Non-rigid Kinematics (SNK), a novel zero-shot method for non-rigid shape matching that eliminates the need for extensive training or ground truth data. SNK operate…

cs.CV2023

Unsupervised Representation Learning for Diverse Deformable Shape Collections

Sara Hahner, Souhaib Attaiki, Jochen Garcke +1

We introduce a novel learning-based method for encoding and manipulating 3D surface meshes. Our method is specifically designed to create an interpretable embedding space for defor…

cs.CV2023

VoroMesh: Learning Watertight Surface Meshes with Voronoi Diagrams

Nissim Maruani, Roman Klokov, Maks Ovsjanikov +2

In stark contrast to the case of images, finding a concise, learnable discrete representation of 3D surfaces remains a challenge. In particular, while polygon meshes are arguably t…

cs.CV2023

Understanding and Improving Features Learned in Deep Functional Maps

Souhaib Attaiki, Maks Ovsjanikov

Deep functional maps have recently emerged as a successful paradigm for non-rigid 3D shape correspondence tasks. An essential step in this pipeline consists in learning feature fun…

cs.CV2023

Generalizable Local Feature Pre-training for Deformable Shape Analysis

Souhaib Attaiki, Lei Li, Maks Ovsjanikov

Transfer learning is fundamental for addressing problems in settings with little training data. While several transfer learning approaches have been proposed in 3D, unfortunately,…