Railgun
Канал о выживании в преддверие зомби апокалипсиса. Меня зовут Вася Шихачевский. Я моушен дизайнер из Москвы и преподаватель в Школе Дизайна ВШЭ. Здесь я пишу о видеоиграх и репощу новости о грядущем киберпанке. И о дизайне еще иногда. @playerfh
نمایش بیشتر- مشترکین
- پوشش پست
- ER - نسبت تعامل
در حال بارگیری داده...
در حال بارگیری داده...
How my avatar learned a backflip 👥 Cr🖤dits and info: VR camera setup in Unity3D: Gerhard Funk instagram.com/g.y.funk Flexible Muscle-Based Locomotion for Bipedal Creatures by Thomas Geijtenbeek, Michiel van de Panne, Frank van der Stappen 🔗
https://www.goatstream.com/research/papers/SA2013/DeepMimic: Example-Guided Deep Reinforcement Learning of Physics-Based Character Skills by Xue Bin Peng, Pieter Abbeel, Sergey Levine, Michiel van de Panne 🔗
https://xbpeng.github.io/projects/DeepMimic/index.htmlCoding: Max Schneider twitter.com/m_schneider 3D Modelling: Leonhard Gläser instagram.com/leonhard.glaeser Voice: Descript Lip and face sync: Mario v. Rickenbach twitter.com/any_user and Cristopher Dorstewitz Sound: David Kamp instagram.com/davidkamp 🔗 backflip:
https://youtu.be/1zUJzzRu-xs?si=gqsoQrCCIU8YjzzhPolycam's Gaussian Splatting demo. Use Gaussian Splatting for free and browse splats published by the community.
A fun film about a topic that scares me a lot. In 2019, I gave backflipping a shot, but it didn't go well. Fear and a broken toe killed my confidence, and I gave up. To get past this, I made an avatar. The training of the avatar was based on the methods from the research paper "Deep Mimic," enabling it to learn to do backflips using machine learning, mimicking human moves, and learning from mistakes without the risk of getting hurt. This journey, full of wonder and discomfort due to A.I.’s potential, took three years to capture, resulting in the above documentary. The film delves into themes like ambition, facing fear, balancing control and uncertainty, aiming for excellence, grappling with the inevitability of technological advancements, accepting failure, and learning to let go. Source: "DeepMimic: Example-Guided Deep Reinforcement Learning of Physics-Based Character Skills" by Xue Bin Peng, Pieter Abbeel, Sergey Levine, Michiel van de Panne 🔗
https://xbpeng.github.io/projects/DeepMimic/index.htmlFilm: Nikita Diakur Code: Maximilian Schneider Sound: David Kamp Music: aiva.ai & Thunderkamp VR cam: Gerhard Funk Camera: Gerhard Funk & María Guadalupe Anaya Alderete Modelling: Leonhard Gläser Production Manager: Karsten Matern Co-Producers: Emmanuel-Alain Raynal, Pierre Baussaron / Miyu Productions Distribution: Luce Gosgean / Miyu Distribution Support: The Open Workshop Artist Residency Funding: MDM, FFA, CNC © Nikita Diakur and Miyu Productions 2022 🔗
https://backflip.training