Fast Rotational Sparse Coding

TL;DR

This paper introduces a fast rotational sparse coding algorithm that incorporates steerability to efficiently handle rotated image features, improving performance in image processing tasks like patch coding and texture classification.

Contribution

It presents a novel rotational sparse coding algorithm based on K-SVD with steerable basis acceleration, enabling practical and efficient handling of rotated image structures.

Findings

Algorithm is fast enough for practical use

Outperforms standard sparse coding in experiments

Effective in texture classification

Abstract

We propose an algorithm for rotational sparse coding along with an efficient implementation using steerability. Sparse coding (also called dictionary learning) is an important technique in image processing, useful in inverse problems, compression, and analysis; however, the usual formulation fails to capture an important aspect of the structure of images: images are formed from building blocks, e.g., edges, lines, or points, that appear at different locations, orientations, and scales. The sparse coding problem can be reformulated to explicitly account for these transforms, at the cost of increased computation. In this work, we propose an algorithm for a rotational version of sparse coding that is based on K-SVD with additional rotation operations. We then propose a method to accelerate these rotations by learning the dictionary in a steerable basis. Our experiments on patch coding and texture classification demonstrate that the proposed algorithm is fast enough for practical use and compares favorably to standard sparse coding.