Computing a Characteristic Orientation for Rotation-Independent Image Analysis

TL;DR

This paper presents GID, a preprocessing technique that estimates and aligns images to a canonical orientation, enhancing rotation robustness in deep learning models without altering their architecture.

Contribution

Introduction of GID, a novel preprocessing method that improves rotation invariance by aligning images to a canonical frame, compatible with standard neural networks.

Findings

GID outperforms state-of-the-art rotation-invariant models on rotated MNIST.

The method maintains effectiveness on complex datasets like CIFAR-10.

GID does not require architectural modifications, reducing computational overhead.

Abstract

Handling geometric transformations, particularly rotations, remains a challenge in deep learning for computer vision. Standard neural networks lack inherent rotation invariance and typically rely on data augmentation or architectural modifications to improve robustness. Although effective, these approaches increase computational demands, require specialised implementations, or alter network structures, limiting their applicability. This paper introduces General Intensity Direction (GID), a preprocessing method that improves rotation robustness without modifying the network architecture. The method estimates a global orientation for each image and aligns it to a canonical reference frame, allowing standard models to process inputs more consistently across different rotations. Unlike moment-based approaches that extract invariant descriptors, this method directly transforms the image while preserving spatial structure, making it compatible with convolutional networks. Experimental evaluation on the rotated MNIST dataset shows that the proposed method achieves higher accuracy than state-of-the-art rotation-invariant architectures. Additional experiments on the CIFAR-10 dataset, confirm that the method remains effective under more complex conditions.