We will present four papers at CVPR 2025 in Nashville: 4Deform: Neural Surface Deformation for Robust Shape Interpolation Higher-Order Ratio Cycles for Fast and Globally Optimal Shape Matching EchoMatch: Partial-to-Partial Shape Matching via Correspondence Reflection Towards Optimizing Large-Scale Multi-Graph Matching in Bioimaging More infos coming soon.

We are looking for multiple outstanding candidates (PhD/Postdoc, fully-funded) in our ERC Starting Grant project (HARMONY) that focuses on cutting-edge research in visual computing (see our press release for more details). The overarching aim of HARMONY is to understand commonalities in visual data (e.g. images, videos, 3D shapes), while relating these to underlying fundamental principles ...

Two papers are conditionally accepted at the ECCV 2024 conference in Milan: Synchronous Diffusion for Unsupervised Smooth Non-Rigid 3D Shape Matching DiscoMatch: Fast Discrete Optimisation for Geometrically Consistent 3D Shape Matching One paper is accepted at the MICCAI 2024 in Marrakesh: A Universal and Flexible Framework for Unsupervised Statistical Shape Model Learning Preprints will be ...

We are honored that our work SpiderMatch: 3D Shape Matching with Global Optimality and Geometric Consistency was selected as best student paper runner-up at CVPR 2024. Check out the article in Computer Vision News.  

We will present four papers at CVPR 2024 in Seattle: – SpiderMatch: 3D Shape Matching with Global Optimality and Geometric Consistency – Spectral Meets Spatial: Harmonising 3D Shape Matching and Interpolation – Partial-to-Partial Shape Matching with Geometric Consistency – Unsupervised 3D Structure Inference from Category-Specific Image Collections

We will present our paper SIGMA: Scale-Invariant Global Sparse Shape Matching at ICCV 2023 in Paris.

Our paper Unsupervised Learning of Robust Spectral Shape Matching has been accepted to SIGGRAPH / TOG.

Two papers at CVPR: Conjugate Product Graphs for Globally Optimal 2D-3D Shape Matching Self-Supervised Learning for Multimodal Non-Rigid 3D Shape Matching

Our paper Universe Points Representation Learning for Partial Multi-Graph Matching has been accepted at AAAI 2023.

Our paper HandFlow: Quantifying View-Dependent 3D Ambiguity in Two-Hand Reconstruction with Normalizing Flow received the Best Paper Honorable Mention Award at VMV 2022.