Facebook’s Prototype Photoreal Avatars No Longer Require Face Tracking
https://youtu.be/OK_BIQPiCEk
https://uploadvr.com/frl-codec-avatars-without-face-tracking/
Training the model requires a multi-camera 3D capture setup with 45 minutes of unique data for each test user. When first shown in 2018, Codec Avatars were described by Facebook as “years away”. While this new research makes the hardware needed to drive avatars more practical, it still doesn’t solve the core issues of generating the avatar in the first place.
#avatars
https://youtu.be/OK_BIQPiCEk
https://uploadvr.com/frl-codec-avatars-without-face-tracking/
Training the model requires a multi-camera 3D capture setup with 45 minutes of unique data for each test user. When first shown in 2018, Codec Avatars were described by Facebook as “years away”. While this new research makes the hardware needed to drive avatars more practical, it still doesn’t solve the core issues of generating the avatar in the first place.
#avatars
YouTube
Facebook's Prototype Photoreal Avatars No Longer Need Face Tracking Cameras
Facebook's photorealistic avatars have come a long way.
Subscribe for the latest VR trailers & clips: https://www.youtube.com/c/VRTrailersClips?sub_confirmation=1
Subscribe for the latest VR trailers & clips: https://www.youtube.com/c/VRTrailersClips?sub_confirmation=1
FaceScape: a Large-scale High Quality 3D Face Dataset and Detailed Riggable 3D Face Prediction
- github
- paper
- project site
- Model example
Features
1. Topologically uniformed.
The geometric models of different identities and different expressions share the same mesh topology, which makes the features on faces easy to be aligned. This also helps in building a 3D morphable model.
2. Displacement map + base mesh.
3. 20 specific expressions.
4. High resolution.
The texture maps and displacement maps reach 4K resolution, which preserving maximum detailed texture and geometry.
#dataset
- github
- paper
- project site
- Model example
Features
1. Topologically uniformed.
The geometric models of different identities and different expressions share the same mesh topology, which makes the features on faces easy to be aligned. This also helps in building a 3D morphable model.
2. Displacement map + base mesh.
3. 20 specific expressions.
4. High resolution.
The texture maps and displacement maps reach 4K resolution, which preserving maximum detailed texture and geometry.
#dataset
https://3dgv.github.io/
3DGV is a virtual seminar series on Geometry Processing and 3D Computer Vision. Seminars take place every Wednesday, cycling through three different time zones (Europe, America, and Asia). We show seminar hours in six different time zones in the following table.
* Youtube
Другие семинары на ютуб на около 3D DL тему
* Matthias Niessner (Technical University of Munich, Visual Computing Lab)
* Vision Seminar MIT
* SU Lab UC San Diego
У последних мне очень понравилась вот эта лекция, которая является неплохим введением в тему 3D DL, а также освещает наиболее значимые работы в классификации, сегментации и генерации 3д объектов.
#seminar #courses
3DGV is a virtual seminar series on Geometry Processing and 3D Computer Vision. Seminars take place every Wednesday, cycling through three different time zones (Europe, America, and Asia). We show seminar hours in six different time zones in the following table.
* Youtube
Другие семинары на ютуб на около 3D DL тему
* Matthias Niessner (Technical University of Munich, Visual Computing Lab)
* Vision Seminar MIT
* SU Lab UC San Diego
У последних мне очень понравилась вот эта лекция, которая является неплохим введением в тему 3D DL, а также освещает наиболее значимые работы в классификации, сегментации и генерации 3д объектов.
#seminar #courses
* https://youtu.be/nkHL1GNU18M
* Very similar video with explanations
https://www.youtube.com/watch?v=0t4aKJuKP0Q
* Very similar video with explanations
https://www.youtube.com/watch?v=0t4aKJuKP0Q
YouTube
Physics in 4 Dimensions…How?
❤️ Check out Weights & Biases and sign up for a free demo here: https://www.wandb.com/papers
❤️ Their mentioned post is available here: https://www.wandb.com/articles/how-to-visualize-models-in-tensorboard-with-weights-and-biases
📝 The paper "N-Dimensional…
❤️ Their mentioned post is available here: https://www.wandb.com/articles/how-to-visualize-models-in-tensorboard-with-weights-and-biases
📝 The paper "N-Dimensional…
Forwarded from PHYGITAL+ART (Oleg Yusupov)
Полезные два сайта для всех, кто интересуется темой AI + искусство
1. https://creative-ai.org - здесь вы найдете большое количество сервисов, доступных для использования без навыков ML
2. https://aiartists.org - большое коммьюнити AI художников для вдохновения
На картине изображена работа Дмитрия Черняка gen2GAN, одна из моих любимых картин совмещающая подходы генеративного искусства и нейронных сетей (в данном случае GAN-нейросети)
1. https://creative-ai.org - здесь вы найдете большое количество сервисов, доступных для использования без навыков ML
2. https://aiartists.org - большое коммьюнити AI художников для вдохновения
На картине изображена работа Дмитрия Черняка gen2GAN, одна из моих любимых картин совмещающая подходы генеративного искусства и нейронных сетей (в данном случае GAN-нейросети)
Art Machines 2: International Symposium on Machine Learning and Art 2021 will take place from 10th-13th June 2021 at the School of Creative Media, City University of Hong Kong.
Art Machines 2 will bring together Academics, Artists and Professionals in the field of Computational Media Art in a four-day symposium on the topic of Machine Learning and Art. This agenda-setting Symposium will provide an opportunity for in depth reflection upon the assessment of the impact of Machine Learning on Artistic Media, as well as to reflect on the conditions of existence, range, futures and social significance of computationally based art and media more generally. In addition to the academic keynotes, panels and discussion forums, the symposium will include a curated exhibition of artworks and a student salon.
https://www.cityu.edu.hk/artmachines2/submission-types
Art Machines 2 will bring together Academics, Artists and Professionals in the field of Computational Media Art in a four-day symposium on the topic of Machine Learning and Art. This agenda-setting Symposium will provide an opportunity for in depth reflection upon the assessment of the impact of Machine Learning on Artistic Media, as well as to reflect on the conditions of existence, range, futures and social significance of computationally based art and media more generally. In addition to the academic keynotes, panels and discussion forums, the symposium will include a curated exhibition of artworks and a student salon.
https://www.cityu.edu.hk/artmachines2/submission-types
www.cityu.edu.hk
Submission Types | Art Machines 2
The open call invites papers and proposals on the following research topics. Please signal which broad area(s) your paper or proposal falls under when you make your submission. All papers should have some relevance to the topic of Artificial Intelligence…
https://neurips2020creativity.github.io/
Machine Learning for Creativity and Design
NeurIPS 2020 Workshop
December 11ᵗʰ-12ᵗʰ
Machine Learning for Creativity and Design
NeurIPS 2020 Workshop
December 11ᵗʰ-12ᵗʰ