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

Abstract : 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 rotation into translation. This space allows the subsequent predictor to learn the internal spatial and angular relations of the objects regardless of their orientation. No data augmentation is needed and the model remains significantly smaller. It presents a self-organization capability and learns to predict the class and the rotation angle without requiring an angle-labeled dataset. We present the results of training with both upright and randomly rotated datasets. The accuracy outperforms the current state of the art on upright oriented training.
Complete list of metadatas

Cited literature [10 references]  Display  Hide  Download

https://hal-enpc.archives-ouvertes.fr/hal-02170933
Contributor : Eva Dokladalova <>
Submitted on : Tuesday, July 2, 2019 - 2:58:42 PM
Last modification on : Tuesday, July 9, 2019 - 1:29:31 AM

File

RotCNN.pdf
Files produced by the author(s)

Identifiers

  • HAL Id : hal-02170933, version 1

Citation

Rosemberg Rodriguez Salas, Petr Dokládal, Eva Dokladalova. RED-NN: Rotation-Equivariant Deep Neural Network for Classification and Prediction of Rotation. 2019. ⟨hal-02170933⟩

Share

Metrics

Record views

36

Files downloads

12