Accelerated Rotation-Invariant Convolution for UAV Image Segmentation

Rotation invariance is essential for precise object-level segmentation in UAV aerial imagery, where targets can have arbitrary orientations and exhibit fine-scale details. Conventional segmentation architectures like UNet rely on convolution operators that are not rotation-invariant, leading to degraded segmentation accuracy across varying viewpoints. Rotation invariance can be achieved by expanding the filter bank across multiple orientations; however, this significantly increases computational cost and memory requirements. In this article, we introduce a Graphics Processing Unit (GPU)-optimized rotation-invariant convolution framework that eliminates the traditional data lowering (im2col) step required for matrix multiplication-based convolution. By exploiting structured data sharing among symmetrically rotated filters, our method achieves multiorientation convolution with greatly reduced memory requirements and computational redundancy. We further generalize the approach to accelerate convolution with arbitrary (nonsymmetric) rotation angles. Integrated into a UNet segmentation model, the framework yields up to a 5.7% improvement in accuracy over the nonrotation-aware baseline. Across extensive benchmarks, the proposed convolution achieves 20%–57% faster training and 15%–45% lower energy consumption than cuDNN, while maintaining accuracy comparable to state-of-the-art rotation-invariant methods. Because the scatter-based operator greatly reduces intermediate feature dimensionality, the efficiency of our design also enables practical 16-orientation convolution and pooling, yielding further accuracy gains that are infeasible for conventional rotation-invariant implementations. Our 16-orientation approach achieves competitive accuracy on multiple datasets compared with state-of-the-art UAV segmentation networks. These results demonstrate that the proposed method provides an effective and efficient alternative to existing rotation-invariant convolution frameworks.