BLADE: Filter Learning for General Purpose Computational Photography

TL;DR

BLADE is a versatile, trainable edge-adaptive filtering framework that generalizes RAISR, enabling efficient solutions for various computational photography tasks like denoising, demosaicing, and stylization.

Contribution

The paper introduces BLADE, a new trainable filtering method that extends RAISR for broader applications in computational photography.

Findings

BLADE achieves efficient image processing in multiple tasks.

It is simple, general, and computationally efficient.

Demonstrates effectiveness in denoising, demosaicing, and stylization.

Abstract

The Rapid and Accurate Image Super Resolution (RAISR) method of Romano, Isidoro, and Milanfar is a computationally efficient image upscaling method using a trained set of filters. We describe a generalization of RAISR, which we name Best Linear Adaptive Enhancement (BLADE). This approach is a trainable edge-adaptive filtering framework that is general, simple, computationally efficient, and useful for a wide range of problems in computational photography. We show applications to operations which may appear in a camera pipeline including denoising, demosaicing, and stylization.