HackerPapers

arXiv:1701.04928cs.CV

Bringing Impressionism to Life with Neural Style Transfer in Come Swim

Bhautik Joshi, Kristen Stewart, David Shapiro

Abstract

Neural Style Transfer is a striking, recently-developed technique that uses neural networks to artistically redraw an image in the style of a source style image. This paper explores the use of this technique in a production setting, applying Neural Style Transfer to redraw key scenes in 'Come Swim' in the style of the impressionistic painting that inspired the film. We document how the technique can be driven within the framework of an iterative creative process to achieve a desired look, and propose a mapping of the broad parameter space to a key set of creative controls. We hope that this mapping can provide insights into priorities for future research.

arXiv comments
3 pages, 6 figures, paper is a case study of how Neural Style Transfer can be used in a movie production context
Journal reference
DOI
Subjects
cs.CVComputer Vision and Pattern Recognition

On Hacker News

Linked by 4 identifiable commenters in 4 comments across 4 threads, from Jan 2017 to Nov 2017. Found in 5 HN submissions from Jan 2017 to Mar 2018. Open submission discussions.

Comments linking this paper, per year.

Discussed in comments4 threads

  1. A Year in Computer Vision 1 comment
    • I might be wrong, and I don't want to question the possibility that someone from a non-technical domain can contribute to technical domains (in particular because neural networks and style transfer are basically high school mathematics; and since this happened before [1]), but…

      erefView on HNParent context

  2. Adobe Sensei 1 comment
    • >I have a strong suspicion Adobe doesn't even know what this product is.......started in a meeting where someone said "all of this exciting research and development around machine learning and visual media" Adobe is doing some really cool work using DL to power is the field of…

      max_View on HNParent context

    • Someone posted a link to the paper on HN a few hours ago: https://news.ycombinator.com/item?id=13443107 Here's the direct link: https://arxiv.org/abs/1701.04928

      mellingView on HNParent context

    • Link to paper authored by Stewart, Bhautik Joshi and David Shapiro of Adobe and Starlight Studios https://arxiv.org/pdf/1701.04928v1.pdf

      kitwalker12View on HNParent context

Submitted to Hacker News

Connected through shared HN threads, not inferred similarity or a recommendation.