AutoStyle: Automatic Style Transfer from Image Collections to Users' Images

Abstract

OriginalNew York Sepia Desert Spring Figure 1: Our system stylizes a user's photo by transferring style from a collection of images returned by a web search for a particular keyword. Examples: sepia (tone of old photos, reduced local contrast), desert (enhanced orange color, reduced global and local contrast), spring (enhanced green color, enhanced saturation, enhanced local contrast), New York (enhanced blue/violet tone, enhanced local contrast). AbstractStylizing photos, to give them an antique or artistic look, has become popular in recent years. The available stylization filters, however, are usually created manually by artists, resulting in a narrow set of choices. Moreover, it can be difficult for the user to select a desired filter, since the filters' names often do not convey their functions. We investigate an approach to photo filtering in which the user provides one or more keywords, and the desired style is defined by the set of images returned by searching the web for those keywords. Our method clusters the returned images, allows the user to select a cluster, then stylizes the user's photos by transferring vignetting, color, and local contrast from that cluster. This approach vastly expands the range of available styles, and gives each filter a meaningful name by default. We demonstrate that our method is able to robustly transfer a wide range of styles from image collections to users' photos.

Real-time user-guided image colorization with learned deep priors

“…Automatic colorization. Early semi-automatic methods (Chia et al 2011;Gupta et al 2012;Irony et al 2005;Liu et al 2008;Welsh et al 2002) utilize an example-based approach that transfers color statistics from a reference image or multiple images (Liu et al 2014;Morimoto et al 2009) to the input grayscale image with techniques such as color transfer (Reinhard et al 2001) and image analogies (Hertzmann et al 2001). These methods work remarkably well when the input and the reference share similar content.…”

Distort-and-Recover: Color Enhancement Using Deep Reinforcement Learning

“…While this approach can provide expressive enhancement and diverse stylizations, the results highly depend on example images while providing proper exemplars is challenging. Recent works [8,7,4] (semi-)automate exemplar selection by image retrieval methods. Liu et al [8] used a keyword-based image search to choose example images.…”

Coloring with Words: Guiding Image Colorization Through Text-Based Palette Generation

“…Since color has a strong association with high-level semantic concepts [10], producing palettes from text input is useful in aiding artists and designers [18] and allows automatic colorization from palettes [42,5]. A downside to using text to choose a filter is that filter names do not usually convey the filter's colors [21], thus making it difficult for users to find the filter that matches their taste just by looking at filter names. To bridge this discrepancy between color palettes and their names, palette recommendation based on user text input has long been studied.…”