WeAreDevelopers Logo WeAreDevelopers
Sign in
For You LIVE
World Congress 2021 • Jun 28, 2021

Deepfakes in Realtime - How Neural Networks Are Changing Our World

Thomas Endres , Martin Förtsch , Jonas Mayer

How do you build real-time deepfakes that rival Hollywood CGI? Discover how autoencoders and GANs conquer live video bottlenecks to instantly swap faces with anatomical accuracy.

Pause
Mute Enter Fullscreen
#1 about 5 min

Introducing deepfakes and the controversy of manipulated media

An overview of public awareness and potential deception surrounding manipulated video content.

#2 about 4 min

Differentiating traditional video fakes from neural network models

How methods like audio-based lip sync, face swap, and parametric face models lack foundational deep learning processing.

#3 about 4 min

Defining true deepfakes generated by trained neural networks

Authentic deepfake generation requires deep learning models like DeepFaceLab to synthesize highly realistic facial replacements.

#4 about 4 min

Preparing video datasets and extracting faces with FaceNet

Extracting and aligning facial landmarks from source videos ensures accurate feature mapping for neural model consumption.

#5 about 5 min

Training autoencoders for latent space facial feature reconstruction

Shared encoders and separate decoders compress and reconstruct facial expressions by leveraging a combined latent space representation.

#6 about 3 min

Implementing the deepfake inference workflow for face generation

The underlying workflow merges the generated facial mask with target video frames using fully trained autoencoder components.

#7 about 5 min

Achieving real-time tracking performance with OpenCV and segmentation

Replacing slow tracking heuristics with OpenCV processes and transfer learning models like MobileNet and U-Net yields rapid full-head segmentation.

#8 about 3 min

Enhancing realistic mask blending through OpenCV image inpainting

Integrating OpenCV image inpainting routines cleanly removes the original head before seamlessly applying the synthetic facial mask.

#9 about 10 min

Stabilizing multiple face generative models using GAN training

Generative adversarial networks counter model deterioration by intelligently adapting the loss balance between a generator and discriminator.

#10 about 3 min

Demonstrating real-time streaming results and target generation constraints

Live showcase examples of head-swapping multiple identities emphasize operational constraints and persistent source limitations.

#11 about 4 min

Exploring media entertainment applications and countering synthetic fakes

Analyzing professional commercial opportunities to replace expensive CGI alongside reviewing an ongoing digital arms race in threat detection.

#12 about 3 min

Future technological advancements in voice cloning and synchronization

Current constraints in architectural pipelines limit deploying sophisticated real-time voice synthesis and accurate full-body physical deepfakes.

Matching moments

5:25 min

Technical pipeline for real-time video deepfakes

Martin Förtsch Martin Förtsch +1 · World Congress 2024

2:47 min

Exploring computer-generated content and modern deepfake capabilities

George Proorocu George Proorocu +1 · World Congress 2024

1:10 min

Analyzing visual artifacts in synthetic media

Martin Förtsch Martin Förtsch +1 · World Congress 2024

2:52 min

Audience questions on deepfake detection, regulations, and ethics

Cheuk Ho · World Congress 2023

1:22 min

Passing technical interviews using active video and audio deepfakes

George Proorocu George Proorocu · World Congress 2026 Europe

2:17 min

Exploiting deepfakes in live video conferencing

Martin Förtsch Martin Förtsch +1 · World Congress 2024

More from World Congress 2021

Micro Frontends with Module Federation: Why and How? thumbnail
45:40
about 5 years ago

Micro Frontends with Module Federation: Why and How?

Manfred Steyer

Data Science on Software Data thumbnail
48:10
about 5 years ago

Data Science on Software Data

Markus Harrer

The journey from developer to devops - what i've learnt along the way thumbnail
42:44
about 5 years ago

The journey from developer to devops - what i've learnt along the way

Liam Hurrel, Alireza Chegini

How Data is Shaping our Games thumbnail
43:30
about 5 years ago

How Data is Shaping our Games

Johanna Pirker

Cross platform Augmented Reality development with React Native thumbnail
40:50
about 5 years ago

Cross platform Augmented Reality development with React Native

Vladimir Novick

How to Build a Monetization Strategy Based on User's in-app Behavior thumbnail
38:59
about 5 years ago

How to Build a Monetization Strategy Based on User's in-app Behavior

Natalie Sieradzki

Upcoming sessions on this topic

Open session

World Congress 2026 North America

September 23, 2026 · 10:00–17:00

Stage 11

Building Stuff with GenAI - The Open Minded Workshop beyond OpenAI

Andreas Erben

CTO for Applied AI and Metaverse at daenet

Andreas Erben
Open session

World Congress 2026 North America

September 24, 2026 · 14:10–14:40

Stage 1

Anatomy of an AI Request: Where Latency and Cost Are Really Born

Dan Fu

VP of Kernels at Together AI

Dan Fu
Open session

World Congress 2026 North America

September 25, 2026 · 11:40–12:10

Stage 9

You Can’t Re-Run Sunlight: Designing ML Data Architectures for Physical AI

An Phan

Senior Data Infrastructure Engineer @ Hippo Harvest

An Phan
Open session

World Congress 2026 North America

September 25, 2026 · 12:55–13:25

Stage 1

From Simulation to Reality: Overcoming the Data Scarcity Crisis in Physical AI

Mitesh Patel

NVIDIA Corporation, Developer Advocate -- Manager

Mitesh Patel
Open session

World Congress 2026 North America

September 24, 2026 · 11:20–11:25

Outdoor Stage

Finding the Edges: Testing, Evaluating, and Monitoring Voice AI Agents Before Your Users Do

Matt Wyman

CEO of Okareo

Matt Wyman
Open session

World Congress 2026 North America

September 23, 2026 · 10:45–12:45

Stage 10

Agents That Own Their Inference: Building Production AI Agents on Dedicated GPUs

Khaja Omer, Sheilah Kirui

Khaja Omer
Sheilah Kirui

Related videos

The AI Elections: How Technology Could Shape Public Sentiment thumbnail
30:02
World Congress 2024 • about 2 years ago

The AI Elections: How Technology Could Shape Public Sentiment

Martin Förtsch, Thomas Endres

AI'll Be Back: Generative AI in Image, Video, and Audio Production thumbnail
29:54
World Congress 2025 • about 1 year ago

AI'll Be Back: Generative AI in Image, Video, and Audio Production

Anonymized By Request, Martin Foertsch, Thomas Endres

Deep Fakes: The Lies We Can’t See thumbnail
24:49
World Congress 2024 • about 2 years ago

Deep Fakes: The Lies We Can’t See

George Proorocu, Mihai Roman

AI Space Factories, Hacking Self-Driving Cars & Detecting Deepfakes thumbnail
3:47
LIVE • 7 months ago

AI Space Factories, Hacking Self-Driving Cars & Detecting Deepfakes

Chris Heilmann, Daniel Cranney

Multimodal Generative AI Demystified thumbnail
56:55
LIVE • almost 3 years ago

Multimodal Generative AI Demystified

Ekaterina Sirazitdinova

Your imaginations is (no longer) the limit: how Generative AI empowers people to be creative thumbnail
29:27
World Congress 2023 • almost 3 years ago

Your imaginations is (no longer) the limit: how Generative AI empowers people to be creative

David Estevez

Related articles

From learning to earning

Jobs that call for the skills explored in this talk.