Uploaded October 2021 | Updated September 2026, 1 week ago
❤️ Become The AI Epiphany
Patreon ❤️ ► patreon.com/theaiepiphany
👨👩👧👦 JOIN OUR DISCORD COMMUNITY:
Discord ► discord.gg/peBrCpheKE
📢 SUBSCRIBE TO MY MONTHLY AI NEWSLETTER:
Substack ► aiepiphany.substack.com
In this video I cover Microsoft's "Fake it till you make it: face analysis in the wild using synthetic data alone" paper.
They introduce synthetic (face) data that is realistic enough to enable trivial generalization to real data!
Disclaimer: I was a part of this team while back at Microsoft, and even had some contributions to this project, but in this video, I only talk about the stuff that can be deduced from the paper itself.
▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬
✅ Paper: arxiv.org/abs/2109.15102
✅ Demo video: youtube.com/watch?v=wlOMpQe8luQ&ab_channel=MicrosoftResearch
▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬
⌚️ Timetable:
00:00 The domain gap problem
04:20 The paper deals only with human faces
06:05 Data-centric approach is short-sighted?
08:50 Building 3D human faces (high-level explanation)
09:30 Pros and cons of the approach
11:45 Building 3D human faces (in-depth)
16:15 Results on landmark localization and sem seg
20:45 Better generalization?
21:25 Outro
▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬
💰 BECOME A PATREON OF THE AI EPIPHANY ❤️
If these videos, GitHub projects, and blogs help you,
consider helping me out by supporting me on Patreon!
The AI Epiphany ► patreon.com/theaiepiphany
One-time donation:
paypal.com/paypalme/theaiepiphany
Much love! ❤️
Huge thank you to these AI Epiphany patreons:
Eli Mahler
Petar Veličković
Bartłomiej Danek
Zvonimir Sabljic
▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬
💡 The AI Epiphany is a channel dedicated to simplifying the field of AI using creative visualizations and in general, a stronger focus on geometrical and visual intuition, rather than the algebraic and numerical "intuition".
▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬
👋 CONNECT WITH ME ON SOCIAL
LinkedIn ► linkedin.com/in/aleksagordic
Twitter ► twitter.com/gordic_aleksa
Instagram ► instagram.com/aiepiphany
Facebook ► facebook.com/aiepiphany
👨👩👧👦 JOIN OUR DISCORD COMMUNITY:
Discord ► discord.gg/peBrCpheKE
📢 SUBSCRIBE TO MY MONTHLY AI NEWSLETTER:
Substack ► aiepiphany.substack.com
💻 FOLLOW ME ON GITHUB FOR ML PROJECTS:
GitHub ► github.com/gordicaleksa
📚 FOLLOW ME ON MEDIUM:
Medium ► gordicaleksa.medium.com
▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬
#facesynthetics #syntheticdata #microsoft
❤️ Become The AI Epiphany
Patreon ❤️ ► patreon.com/theaiepiphany
👨👩👧👦 JOIN OUR DISCORD COMMUNITY:
Discord ► discord.gg/peBrCpheKE
📢 SUBSCRIBE TO MY MONTHLY AI NEWSLETTER:
Substack ► aiepiphany.substack.com
In this video I cover Microsoft's "Fake it till you make it: face analysis in the wild using synthetic data alone" paper.
They introduce synthetic (face) data that is realistic enough to enable trivial generalization to real data!
Disclaimer: I was a part of this team while back at Microsoft, and even had some contributions to this project, but in this video, I only talk about the stuff that can be deduced from the paper itself.
▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬
✅ Paper: arxiv.org/abs/2109.15102
✅ Demo video: youtube.com/watch?v=wlOMpQe8luQ&ab_channel=MicrosoftResearch
▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬
⌚️ Timetable:
00:00 The domain gap problem
04:20 The paper deals only with human faces
06:05 Data-centric approach is short-sighted?
08:50 Building 3D human faces (high-level explanation)
09:30 Pros and cons of the approach
11:45 Building 3D human faces (in-depth)
16:15 Results on landmark localization and sem seg
20:45 Better generalization?
21:25 Outro
▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬
💰 BECOME A PATREON OF THE AI EPIPHANY ❤️
If these videos, GitHub projects, and blogs help you,
consider helping me out by supporting me on Patreon!
The AI Epiphany ► patreon.com/theaiepiphany
One-time donation:
paypal.com/paypalme/theaiepiphany
Much love! ❤️
Huge thank you to these AI Epiphany patreons:
Eli Mahler
Petar Veličković
Bartłomiej Danek
Zvonimir Sabljic
▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬
💡 The AI Epiphany is a channel dedicated to simplifying the field of AI using creative visualizations and in general, a stronger focus on geometrical and visual intuition, rather than the algebraic and numerical "intuition".
▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬
👋 CONNECT WITH ME ON SOCIAL
LinkedIn ► linkedin.com/in/aleksagordic
Twitter ► twitter.com/gordic_aleksa
Instagram ► instagram.com/aiepiphany
Facebook ► facebook.com/aiepiphany
👨👩👧👦 JOIN OUR DISCORD COMMUNITY:
Discord ► discord.gg/peBrCpheKE
📢 SUBSCRIBE TO MY MONTHLY AI NEWSLETTER:
Substack ► aiepiphany.substack.com
💻 FOLLOW ME ON GITHUB FOR ML PROJECTS:
GitHub ► github.com/gordicaleksa
📚 FOLLOW ME ON MEDIUM:
Medium ► gordicaleksa.medium.com
▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬
#facesynthetics #syntheticdata #microsoft
