{"about":{"site":"https://codewithpapers.app","non_affiliation":"Code with Papers and Syntology are not affiliated with, endorsed by, or sponsored by Papers with Code, Meta, or the pwc-archive mirror.","licence":"CC BY-SA 4.0","licence_url":"https://creativecommons.org/licenses/by-sa/4.0/legalcode","attribution":"https://codewithpapers.app/attribution","modified":"archive material modified by Syntology; see the attribution page"},"url":"/paper/towards-metamerism-via-foveated-style","title":"Towards Metamerism via Foveated Style Transfer","arxiv_id":"1705.10041","date":"2017-05-29","proceeding":"ICLR 2019 5","authors":["Arturo Deza","Aditya Jonnalagadda","Miguel Eckstein"],"abstract":"The problem of $\\textit{visual metamerism}$ is defined as finding a family of\nperceptually indistinguishable, yet physically different images. In this paper,\nwe propose our NeuroFovea metamer model, a foveated generative model that is\nbased on a mixture of peripheral representations and style transfer\nforward-pass algorithms. Our gradient-descent free model is parametrized by a\nfoveated VGG19 encoder-decoder which allows us to encode images in high\ndimensional space and interpolate between the content and texture information\nwith adaptive instance normalization anywhere in the visual field. Our\ncontributions include: 1) A framework for computing metamers that resembles a\nnoisy communication system via a foveated feed-forward encoder-decoder network\n-- We observe that metamerism arises as a byproduct of noisy perturbations that\npartially lie in the perceptual null space; 2) A perceptual optimization scheme\nas a solution to the hyperparametric nature of our metamer model that requires\ntuning of the image-texture tradeoff coefficients everywhere in the visual\nfield which are a consequence of internal noise; 3) An ABX psychophysical\nevaluation of our metamers where we also find that the rate of growth of the\nreceptive fields in our model match V1 for reference metamers and V2 between\nsynthesized samples. Our model also renders metamers at roughly a second,\npresenting a $\\times1000$ speed-up compared to the previous work, which allows\nfor tractable data-driven metamer experiments.","url_abs":"http://arxiv.org/abs/1705.10041v3","url_pdf":"http://arxiv.org/pdf/1705.10041v3.pdf","source":{"archive":"pwc-archive (Hugging Face), CC BY-SA 4.0","snapshot":"2025-07-28","licence_url":"https://creativecommons.org/licenses/by-sa/4.0/legalcode","row_kind":"abstracts"},"code_links":[{"paper_slug":"towards-metamerism-via-foveated-style","repo_url":"https://github.com/ArturoDeza/NeuroFovea","is_official":1,"mentioned_in_paper":0,"mentioned_in_github":0,"framework":"pytorch","reach":{"status":"unanswered"}}],"tasks":[{"task_slug":"decoder","task_name":"Decoder"},{"task_slug":"metamerism","task_name":"Metamerism"},{"task_slug":"style-transfer","task_name":"Style Transfer"},{"task_slug":"texture-synthesis","task_name":"Texture Synthesis"}],"methods":[{"method_slug":"adaptive-instance-normalization","method_name":"Adaptive Instance Normalization"}],"datasets_introduced":[],"methods_introduced":[],"results":[],"syntology":{"syntology_url":"https://syntology.ai/paper/1705.10041","atlas_url":"https://app.syntology.ai/?focus=1705.10041","mcp":null,"developers":"https://syntology.ai/developers"},"arxiv_metadata":null,"syntology_extracted_results":null}