Rotation and Translation Invariant Representation Learning with Implicit Neural Representations

Sehyun Kwon, Joo Young Choi, Ernest K. Ryu
Proceedings of the 40th International Conference on Machine Learning, PMLR 202:18037-18056, 2023.

Abstract

In many computer vision applications, images are acquired with arbitrary or random rotations and translations, and in such setups, it is desirable to obtain semantic representations disentangled from the image orientation. Examples of such applications include semiconductor wafer defect inspection, plankton microscope images, and inference on single-particle cryo-electron microscopy (cryo-EM) micro-graphs. In this work, we propose Invariant Representation Learning with Implicit Neural Representation (IRL-INR), which uses an implicit neural representation (INR) with a hypernetwork to obtain semantic representations disentangled from the orientation of the image. We show that IRL-INR can effectively learn disentangled semantic representations on more complex images compared to those considered in prior works and show that these semantic representations synergize well with SCAN to produce state-of-the-art unsupervised clustering results.

Cite this Paper

BibTeX
@InProceedings{pmlr-v202-kwon23a, title = {Rotation and Translation Invariant Representation Learning with Implicit Neural Representations}, author = {Kwon, Sehyun and Choi, Joo Young and Ryu, Ernest K.}, booktitle = {Proceedings of the 40th International Conference on Machine Learning}, pages = {18037--18056}, year = {2023}, editor = {Krause, Andreas and Brunskill, Emma and Cho, Kyunghyun and Engelhardt, Barbara and Sabato, Sivan and Scarlett, Jonathan}, volume = {202}, series = {Proceedings of Machine Learning Research}, month = {23--29 Jul}, publisher = {PMLR}, pdf = {https://proceedings.mlr.press/v202/kwon23a/kwon23a.pdf}, url = {https://proceedings.mlr.press/v202/kwon23a.html}, abstract = {In many computer vision applications, images are acquired with arbitrary or random rotations and translations, and in such setups, it is desirable to obtain semantic representations disentangled from the image orientation. Examples of such applications include semiconductor wafer defect inspection, plankton microscope images, and inference on single-particle cryo-electron microscopy (cryo-EM) micro-graphs. In this work, we propose Invariant Representation Learning with Implicit Neural Representation (IRL-INR), which uses an implicit neural representation (INR) with a hypernetwork to obtain semantic representations disentangled from the orientation of the image. We show that IRL-INR can effectively learn disentangled semantic representations on more complex images compared to those considered in prior works and show that these semantic representations synergize well with SCAN to produce state-of-the-art unsupervised clustering results.} }
Endnote
%0 Conference Paper %T Rotation and Translation Invariant Representation Learning with Implicit Neural Representations %A Sehyun Kwon %A Joo Young Choi %A Ernest K. Ryu %B Proceedings of the 40th International Conference on Machine Learning %C Proceedings of Machine Learning Research %D 2023 %E Andreas Krause %E Emma Brunskill %E Kyunghyun Cho %E Barbara Engelhardt %E Sivan Sabato %E Jonathan Scarlett %F pmlr-v202-kwon23a %I PMLR %P 18037--18056 %U https://proceedings.mlr.press/v202/kwon23a.html %V 202 %X In many computer vision applications, images are acquired with arbitrary or random rotations and translations, and in such setups, it is desirable to obtain semantic representations disentangled from the image orientation. Examples of such applications include semiconductor wafer defect inspection, plankton microscope images, and inference on single-particle cryo-electron microscopy (cryo-EM) micro-graphs. In this work, we propose Invariant Representation Learning with Implicit Neural Representation (IRL-INR), which uses an implicit neural representation (INR) with a hypernetwork to obtain semantic representations disentangled from the orientation of the image. We show that IRL-INR can effectively learn disentangled semantic representations on more complex images compared to those considered in prior works and show that these semantic representations synergize well with SCAN to produce state-of-the-art unsupervised clustering results.
APA
Kwon, S., Choi, J.Y. & Ryu, E.K.. (2023). Rotation and Translation Invariant Representation Learning with Implicit Neural Representations. Proceedings of the 40th International Conference on Machine Learning, in Proceedings of Machine Learning Research 202:18037-18056 Available from https://proceedings.mlr.press/v202/kwon23a.html.

Related Material