Corpus-based visual synthesis

In this paper, we present a novel method for creating a painted image from a photograph using an existing painting as a style source. The core idea is to identify the corresponding objects in the two images in order to select patches more appropriately. We automatically make a region correspondence between the painted source image and the target photograph by computing color and texture feature distances. Next, we conduct a patch-based synthesis that preserves the appropriate source and target features. Unlike previous example-based approaches of painting style transfer, our results successfully reflect the features of the source images even if the input images have various colors and textures. Our method allows us to automatically render a new painted image preserving the features of the source image.

show abstract

Autoencoding Blade Runner

Broad

Grierson

2017

ACM SIGGRAPH 2017 Art Gallery

View full text Add to dashboard Cite

Blade Runner-Autoencoded' is a film made by training an autoencoder-a type of generative neural network-to recreate frames from the film Blade Runner. The autoencoder is made to reinterpret every individual frame, reconstructing it based on its memory of the film. The result is a hazy, dreamlike version of the original film. The project explores the aesthetic qualities of the disembodied gaze of the neural network. The autoencoder is also capable of representing images from films it has not seen based on what it has learned from watching Blade Runner. Introduction Reconstructing videos based on prior visual information has some scientific and artistic precedents. Casey and Grierson [1] present a system for real-time matching of an audio input stream to a database of continuous audio or video, presenting an application called REMIX-TV. Grierson develops on this work with PLUNDERMATICS [2], adding more sophisticated methods for feature extraction, segmentation and filtering. Mital, Grierson, and Smith [3] extend this approach further to synthesis a target image using a corpus of images. The image is synthesised in fragments that are matched from the database extracted from the corpus based on shape and colour similarity. Mital uses this technique to create a series of artworks called 'YouTube Smash Up' [4], synthesising the week's most popular video on YouTube from fragments of other trending videos on the platform. Another, somewhat related approach (and key influence to this project) is the research in reconstructing what people are watching while in an MRI scanner, solely from recorded brain scans [5].

show abstract

scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.

Contact Info

customersupport@researchsolutions.com

10624 S. Eastern Ave., Ste. A-614

Henderson, NV 89052, USA

This site is protected by reCAPTCHA and the Google Privacy Policy and Terms of Service apply.

Blog Terms and Conditions API Terms Privacy Policy Contact Cookie Preferences Do Not Sell or Share My Personal Information

Made with 💙 for researchers

Part of the Research Solutions Family.

Corpus-based visual synthesis

Cited by 4 publications

References 27 publications

Audiovisual Resynthesis in an Augmented Reality

Audiovisual Resynthesis in an Augmented Reality

Region-based painting style transfer

Autoencoding Blade Runner

Contact Info

Product

Resources

About