Learning shape correspondence with anisotropic convolutional neural networks

Davide Boscaini, Jonathan Masci, Emanuele Rodolà, Michael Bronstein

Research output: Contribution to journalConference articlepeer-review


Convolutional neural networks have achieved extraordinary results in many computer vision and pattern recognition applications; however, their adoption in the computer graphics and geometry processing communities is limited due to the non-Euclidean structure of their data. In this paper, we propose Anisotropic Convolutional Neural Network (ACNN), a generalization of classical CNNs to non-Euclidean domains, where classical convolutions are replaced by projections over a set of oriented anisotropic diffusion kernels. We use ACNNs to effectively learn intrinsic dense correspondences between deformable shapes, a fundamental problem in geometry processing, arising in a wide variety of applications. We tested ACNNs performance in challenging settings, achieving state-of-the-art results on recent correspondence benchmarks.

Original languageEnglish
Pages (from-to)3197-3205
Number of pages9
JournalAdvances in Neural Information Processing Systems
StatePublished - 2016
Event30th Annual Conference on Neural Information Processing Systems, NIPS 2016 - Barcelona, Spain
Duration: 5 Dec 201610 Dec 2016


FundersFunder number
ERC Starting307047


    Dive into the research topics of 'Learning shape correspondence with anisotropic convolutional neural networks'. Together they form a unique fingerprint.

    Cite this