Barakeel Fanseu Kamhoua — accepted AI-conference papers

2024

HOPE: Shape Matching Via Aligning Different K-hop Neighbourhoods

NeurIPS 2024poster

Accurate and smooth shape matching is very hard to achieve. This is because for accuracy, one needs unique descriptors (signatures) on shapes that distinguish different vertices on a mesh accurately while at the same time being invariant to deformations. However, most existing unique shape descripto…

Cited by 0 Source ↗PDF ↗Scholar ↗

2022

Exact Shape Correspondence via 2D graph convolution

Barakeel Fanseu Kamhoua, Lin Zhang, Yongqiang Chen, Han Yang, MA KAILI, Bo Han +2

NeurIPS 2022accept

For exact 3D shape correspondence (matching or alignment), i.e., the task of matching each point on a shape to its exact corresponding point on the other shape (or to be more specific, matching at geodesic error 0), most existing methods do not perform well due to two main problems. First, on nearly…

Cited by 6 Source ↗PDF ↗Scholar ↗