Neural Feature Matching in Implicit 3D Representations

Yunlu Chen, Basura Fernando, Hakan Bilen, Thomas Mensink, Efstratios Gavves
Proceedings of the 38th International Conference on Machine Learning, PMLR 139:1582-1593, 2021.

Abstract

Recently, neural implicit functions have achieved impressive results for encoding 3D shapes. Conditioning on low-dimensional latent codes generalises a single implicit function to learn shared representation space for a variety of shapes, with the advantage of smooth interpolation. While the benefits from the global latent space do not correspond to explicit points at local level, we propose to track the continuous point trajectory by matching implicit features with the latent code interpolating between shapes, from which we corroborate the hierarchical functionality of the deep implicit functions, where early layers map the latent code to fitting the coarse shape structure, and deeper layers further refine the shape details. Furthermore, the structured representation space of implicit functions enables to apply feature matching for shape deformation, with the benefits to handle topology and semantics inconsistency, such as from an armchair to a chair with no arms, without explicit flow functions or manual annotations.

Cite this Paper

BibTeX
@InProceedings{pmlr-v139-chen21f, title = {Neural Feature Matching in Implicit 3D Representations}, author = {Chen, Yunlu and Fernando, Basura and Bilen, Hakan and Mensink, Thomas and Gavves, Efstratios}, booktitle = {Proceedings of the 38th International Conference on Machine Learning}, pages = {1582--1593}, year = {2021}, editor = {Meila, Marina and Zhang, Tong}, volume = {139}, series = {Proceedings of Machine Learning Research}, month = {18--24 Jul}, publisher = {PMLR}, pdf = {http://proceedings.mlr.press/v139/chen21f/chen21f.pdf}, url = {https://proceedings.mlr.press/v139/chen21f.html}, abstract = {Recently, neural implicit functions have achieved impressive results for encoding 3D shapes. Conditioning on low-dimensional latent codes generalises a single implicit function to learn shared representation space for a variety of shapes, with the advantage of smooth interpolation. While the benefits from the global latent space do not correspond to explicit points at local level, we propose to track the continuous point trajectory by matching implicit features with the latent code interpolating between shapes, from which we corroborate the hierarchical functionality of the deep implicit functions, where early layers map the latent code to fitting the coarse shape structure, and deeper layers further refine the shape details. Furthermore, the structured representation space of implicit functions enables to apply feature matching for shape deformation, with the benefits to handle topology and semantics inconsistency, such as from an armchair to a chair with no arms, without explicit flow functions or manual annotations.} }
Endnote
%0 Conference Paper %T Neural Feature Matching in Implicit 3D Representations %A Yunlu Chen %A Basura Fernando %A Hakan Bilen %A Thomas Mensink %A Efstratios Gavves %B Proceedings of the 38th International Conference on Machine Learning %C Proceedings of Machine Learning Research %D 2021 %E Marina Meila %E Tong Zhang %F pmlr-v139-chen21f %I PMLR %P 1582--1593 %U https://proceedings.mlr.press/v139/chen21f.html %V 139 %X Recently, neural implicit functions have achieved impressive results for encoding 3D shapes. Conditioning on low-dimensional latent codes generalises a single implicit function to learn shared representation space for a variety of shapes, with the advantage of smooth interpolation. While the benefits from the global latent space do not correspond to explicit points at local level, we propose to track the continuous point trajectory by matching implicit features with the latent code interpolating between shapes, from which we corroborate the hierarchical functionality of the deep implicit functions, where early layers map the latent code to fitting the coarse shape structure, and deeper layers further refine the shape details. Furthermore, the structured representation space of implicit functions enables to apply feature matching for shape deformation, with the benefits to handle topology and semantics inconsistency, such as from an armchair to a chair with no arms, without explicit flow functions or manual annotations.
APA
Chen, Y., Fernando, B., Bilen, H., Mensink, T. & Gavves, E.. (2021). Neural Feature Matching in Implicit 3D Representations. Proceedings of the 38th International Conference on Machine Learning, in Proceedings of Machine Learning Research 139:1582-1593 Available from https://proceedings.mlr.press/v139/chen21f.html.

Related Material