Computing a Characteristic Orientation for Rotation-Independent Image Analysis
This addresses the problem of computational inefficiency and architectural limitations in achieving rotation invariance for computer vision practitioners, though it is incremental as it builds on existing preprocessing ideas.
The paper tackles the challenge of rotation invariance in deep learning for computer vision by introducing General Intensity Direction (GID), a preprocessing method that aligns images to a canonical orientation without modifying network architectures, achieving higher accuracy than state-of-the-art rotation-invariant methods on rotated MNIST and remaining effective on CIFAR-10.
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.