Achieving Rotation Invariance in Convolution Operations: Shifting from Data-Driven to Mechanism-Assured
This work addresses the challenge of rotation invariance in computer vision tasks, such as object recognition and remote sensing, by providing a mechanism-assured approach that enhances model performance, particularly in data-scarce scenarios.
The paper tackled the problem of achieving rotation invariance in deep neural networks without relying on data by designing new convolution operations (RIConvs) that are naturally invariant to arbitrary rotations, resulting in state-of-the-art performance on tasks like multi-orientation object recognition and improved accuracy in texture recognition, aircraft type recognition, and remote sensing image classification, especially with limited training data.
Achieving rotation invariance in deep neural networks without relying on data has always been a hot research topic. Intrinsic rotation invariance can enhance the model's feature representation capability, enabling better performance in tasks such as multi-orientation object recognition and detection. Based on various types of non-learnable operators, including gradient, sort, local binary pattern, maximum, etc., this paper designs a set of new convolution operations that are natually invariant to arbitrary rotations. Unlike most previous studies, these rotation-invariant convolutions (RIConvs) have the same number of learnable parameters and a similar computational process as conventional convolution operations, allowing them to be interchangeable. Using the MNIST-Rot dataset, we first verify the invariance of these RIConvs under various rotation angles and compare their performance with previous rotation-invariant convolutional neural networks (RI-CNNs). Two types of RIConvs based on gradient operators achieve state-of-the-art results. Subsequently, we combine RIConvs with different types and depths of classic CNN backbones. Using the OuTex_00012, MTARSI, and NWPU-RESISC-45 datasets, we test their performance on texture recognition, aircraft type recognition, and remote sensing image classification tasks. The results show that RIConvs significantly improve the accuracy of these CNN backbones, especially when the training data is limited. Furthermore, we find that even with data augmentation, RIConvs can further enhance model performance.