Learning to Remove Soft Shadows

• DOI: 10.1145/2732407
• Publication Date: 2015-11-05T16:19:01Z
• Authors: Maciej Gryka; Michael Terry; Gabriel J. Brostow
• Keywords: Shadow mapping

Manipulated images lose believability if the user's edits fail to account for shadows. We propose a method that makes removal and editing of soft shadows easy. Soft are ubiquitous, but remain notoriously difficult extract manipulate. posit can be segmented, therefore edited, by learning mapping function image patches generates shadow mattes. validate this premise removing from photographs with only small amount user input. Given broad brush strokes indicate region processed, our new supervised regression algorithm automatically unshadows an image, umbra penumbra. The resulting lit is frequently perceived as believable shadow-free version scene. tested approach on large set images, performed study compared state-of-the-art real scenes. Our results more identify being altered preferable prior work.