# Real-Time Holography Simulation! ## Метаданные - **Канал:** Two Minute Papers - **YouTube:** https://www.youtube.com/watch?v=duCQUu8EQVA - **Дата:** 31.10.2018 - **Длительность:** 3:05 - **Просмотры:** 33,575 ## Описание The paper "Acquiring Spatially Varying Appearance of Printed Holographic Surfaces" is available here: https://wp.doc.ic.ac.uk/rgi/project/acquiring-spatially-varying-appearance-of-printed-holographic-surfaces/ Material learning and synthesis algorithm: https://users.cg.tuwien.ac.at/zsolnai/gfx/gaussian-material-synthesis/ Neural Material Synthesis: https://www.youtube.com/watch?v=XpwW3glj2T8 Pick up cool perks on our Patreon page: › https://www.patreon.com/TwoMinutePapers We would like to thank our generous Patreon supporters who make Two Minute Papers possible: 313V, Andrew Melnychuk, Angelos Evripiotis, Anthony Vdovitchenko, Brian Gilman, Christian Ahlin, Christoph Jadanowski, Dennis Abts, Emmanuel, Eric Haddad, Eric Martel, Evan Breznyik, Geronimo Moralez, John De Witt, Kjartan Olason, Lorin Atzberger, Marten Rauschenberg, Michael Albrecht, Michael Jensen, Milan Lajtoš, Morten Punnerud Engelstad, Nader Shakerin, Owen Skarpness, Raul Araújo da Silva, Rob Rowe, Robin Graham, Ryan Monsurate, Shawn Azman, Steef, Steve Messina, Sunil Kim, Thomas Krcmar, Torsten Reil, Zach Boldyga. https://www.patreon.com/TwoMinutePapers Two Minute Papers Merch: US: https://www.amazon.com/dp/B06ZZ23V2L EU/Worldwide: https://shop.spreadshirt.net/TwoMinutePapers/ Thumbnail background image credit: https://pixabay.com/photo-1357029/ Splash screen/thumbnail design: Felícia Fehér - http://felicia.hu Károly Zsolnai-Fehér's links: Facebook: https://www.facebook.com/TwoMinutePapers/ Twitter: https://twitter.com/karoly_zsolnai Web: https://cg.tuwien.ac.at/~zsolnai/ ## Содержание ### [0:00](https://www.youtube.com/watch?v=duCQUu8EQVA) Segment 1 (00:00 - 03:00) Dear Fellow Scholars, this is Two Minute Papers with Károly Zsolnai-Fehér. If we wish to populate a virtual world with photorealistic materials, the last few years have offered a number of amazing techniques to do so. We can obtain such a material from a flash and no-flash photograph pair of the target material, and having a neural network create a digital version of it, or remarkably, even just one photograph is enough to perform this. This footage that you see here shows these materials after they have been rendered by a light simulation program. If we don't have physical access to these materials, we can also use a recent learning algorithm to learn our preferences and recommend new materials that we would enjoy. However, whenever I publish such a video, I always get comments asking: "but what about the more advanced materials? ", and my answer is, you are right! Have a look at this piece of work which is about acquiring printed holographic materials. This means that we have physical access to this holographic pattern, put a camera close by, and measure data in a way that can be imported into a light simulation program to make a digital copy of it. This idea is much less far-fetched than it sounds, because we find such materials in many everyday objects, like bank notes, gift bags, clothing, or of course or security holograms. However, it is also quite difficult. Look here! As you see, these holographic patterns are quite diverse, and a well-crafted algorithm would have to be able to capture this rotation effect, circular diffractive areas, firework effects and even this iridescent glitter. That is quite a challenge! This paper proposes two novel techniques to approach this problem. The first one assumes that there is some sort of repetition in the visual structure of the hologram and takes that into consideration. As a result, it can give us high-quality results by only taking 1 to 5 photographs of the target material. The second method is more exhaustive and needs more specialized hardware, but, in return, can deal with arbitrary structures, and requires at least 4 photographs at all times. These are both quite remarkable - just think about the fact that these materials look different from every viewing angle, and they also change over the surface of the object. And for the first technique, we don't need sophisticated instruments, only a consumer DSLR camera is required. The reconstructed digital materials can be used in real time, and what's more, we can also exert artistic control over the outputs by modifying the periodicities of the material. How cool is that! And, if you are about to subscribe to the series or you are already subscribed, make sure to click the bell icon, or otherwise, you may miss future episodes. That would be a bummer, because I have a lot more amazing papers to show you. Thanks for watching and for your generous support, and I'll see you next time! --- *Источник: https://ekstraktznaniy.ru/video/14397*