Graph-Based Classification of Omnidirectional Images
This addresses the challenge of non-optimal solutions in robotics and virtual reality due to classical methods designed for planar images, though it appears incremental as it extends deep learning to graphs for this domain.
The paper tackled the problem of classifying omnidirectional images by accounting for their specific geometry using graph-based representations, and the proposed method outperformed current techniques.
Omnidirectional cameras are widely used in such areas as robotics and virtual reality as they provide a wide field of view. Their images are often processed with classical methods, which might unfortunately lead to non-optimal solutions as these methods are designed for planar images that have different geometrical properties than omnidirectional ones. In this paper we study image classification task by taking into account the specific geometry of omnidirectional cameras with graph-based representations. In particular, we extend deep learning architectures to data on graphs; we propose a principled way of graph construction such that convolutional filters respond similarly for the same pattern on different positions of the image regardless of lens distortions. Our experiments show that the proposed method outperforms current techniques for the omnidirectional image classification problem.