“A Gentle Introduction to Bilateral Filtering and its Applications” by Paris, Kornprobst, Tumblin and Durand

  • ©Sylvain Paris, Pierre Kornprobst, Jack Tumblin, and Frédo Durand


Entry Number: 05


    A Gentle Introduction to Bilateral Filtering and its Applications



    We welcome all motivated beginners. Don’t worry if integrals or Gaussian functions are unfamiliar: we only assume that attendees know some digital image basics (pixels, gray levels, noise) and have some modest programming experience. If you can compute a weighted average, then you are ready to take this course.

    Who Should Attend
    This course is meant to introduce a graphics person to bilateral filtering. We typically target someone who wants to get into image editing, computational photography, and other related fields. We cover issues that appeal to anyone interested in image manipulations: researchers, developers, software designers and more.

    The bilateral filter is a nonlinear process that smooths an image while preserving its edges. Although this description may sound scary to most of us, this filter is nothing other than a basic weighted average. This simple tool has become ubiquitous in image processing and has shown remarkable abilities to filter images, videos, and even 3D meshes. Our course presents its most successful applications, describes its various implementations, and exposes in a comprehensive manner the related theoretical background. The attendees will learn progressively many basic concepts underpinning bilateral filtering. The course starts with classical Gaussian blur, then defines the bilateral filter, and ends with extensions such as cross bilateral filtering. These discussions are systematically complemented with demonstrations of state-of-the-art applications such as tone mapping and photograph enhancement. Finally, we show that all the benefits of bilateral filtering are contained within a few dozen lines of code.

    The bilateral filter is ubiquitous in computational photography applications. It is increasingly common in computer graphics research papers but no single reference summarizes its properties and applications. This course provides a graphical, strongly intuitive introduction to bilateral filtering, and a practical guide for image editing, tone-maps, video processing and more.