通过预处理对齐图像方向，让普通模型也能抗旋转干扰。
Computing a Characteristic Orientation for Rotation-Independent Image Analysis
- 为每张图计算全局方向并统一到标准朝向
- 在旋转MNIST上准确率超过现有旋转不变模型
- 无需改网络结构，适合快速提升现有模型性能
处理几何变换，尤其是旋转，仍是计算机视觉深度学习中的挑战。标准神经网络缺乏固有的旋转不变性，通常依赖数据增强或结构修改来提升鲁棒性。尽管有效，这些方法会增加计算开销、需要特殊实现或改变网络结构，限制了应用范围。本文提出通用强度方向（GID）预处理方法，在不修改网络结构的前提下提升旋转鲁棒性。该方法为每张图像估计全局方向，并将其对齐至标准参考坐标系，使标准模型能更一致地处理不同旋转的输入。不同于基于矩的不变描述子提取方法，该方法直接变换图像，同时保留空间结构，与卷积网络兼容。在旋转MNIST数据集上的实验表明，该方法达到比现有先进旋转不变架构更高的准确率。在更复杂的CIFAR-10数据集上的额外实验也验证了其有效性。
原文摘要 · Abstract (English)
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.
Thank you to arXiv for use of its open access interoperability. PaperDance 不是 arXiv 官方产品；中文卡片由大模型生成，请以原文为准。