A minimal model for rotation invariant convolutional neural networks with angle prediction

Autor: Rosemberg Rodriguez Salas, Petr Dokladal, Eva Dokladalova

Publikováno v: 16th International Conference on Computer Vision Theory and Applications-VISAPP 2021
16th International Conference on Computer Vision Theory and Applications-VISAPP 2021, Feb 2021
Streaming, Austria
HAL

International audience

Externí odkaz: https://explore.openaire.eu/search/publication?articleId=dedup_wf_001::42b3382403dd24fde80e340f1f9fea29
https://hal-upec-upem.archives-ouvertes.fr/hal-03111509

RED-NN: Rotation-Equivariant Deep Neural Network for Classification and Prediction of Rotation

Autor: Rosemberg Rodriguez Salas, Petr Dokladal, Eva Dokladalova

Publikováno v: HAL

In this work, we propose a new Convolutional Neural Network (CNN) for classification of rotated objects. This architecture is built around an ordered ensemble of oriented edge detectors to create a roto-translational space that transforms the input r

Externí odkaz: https://explore.openaire.eu/search/publication?articleId=dedup_wf_001::91372415975f453f6370815dfdbe8e09
https://hal-enpc.archives-ouvertes.fr/hal-02170933/file/RotCNN.pdf

A minimal model for classification of rotated objects with prediction of the angle of rotation

Autor: Rosemberg Rodriguez Salas, Petr Dokládal, Eva Dokladalova

Publikováno v: Journal of Visual Communication and Image Representation
Journal of Visual Communication and Image Representation, Elsevier, 2021, 75, pp.103054. ⟨10.1016/j.jvcir.2021.103054⟩

International audience; In classification tasks, the robustness against various image transformations remains a crucial property of the Convolutional Neural Networks (CNNs). It can be acquired using the data augmentation. It comes, however, at the pr

Externí odkaz: https://explore.openaire.eu/search/publication?articleId=doi_dedup___::1b455f89031a4f4df725a76727c25df6
https://doi.org/10.1016/j.jvcir.2021.103054