Sony Pixel Power calrec Sony

More Than Meets the AI: How GANs Research Is Reshaping Video Conferencing

24/06/2021

Roll out of bed, fire up the laptop, turn on the webcam - and look picture-perfect in every video call, with the help of AI developed by NVIDIA researchers.

Vid2Vid Cameo, one of the deep learning models behind the NVIDIA Maxine software development kit for video conferencing, uses generative adversarial networks (known as GANs) to synthesize realistic talking-head videos using a single 2D image of a person.

To use it, participants submit a reference image - which could be either a real photo of themselves or a cartoon avatar - before joining a video call. During the meeting, the AI model will capture each individual's real-time motion and apply it to the previously uploaded still image.

That means that by uploading a photo of themselves in formal attire, meeting attendees with mussed hair and pajamas can appear on a call in work-appropriate attire, with AI mapping the user's facial movements to the reference photo. If the subject is turned to the left, the technology can adjust the viewpoint so the attendee appears to be directly facing the webcam.

Besides helping meeting attendees look their best, this AI technique also shrinks the bandwidth needed for video conferencing by up to 10x, avoiding jitter and lag. It'll soon be available in the NVIDIA Video Codec SDK as the AI Face Codec.

Many people have limited internet bandwidth, but still want to have a smooth video call with friends and family, said NVIDIA researcher Ming-Yu Liu, co-author on the project. In addition to helping them, the underlying technology could also be used to assist the work of animators, photo editors and game developers.

Vid2Vid Cameo was presented this week at the prestigious Conference on Computer Vision and Pattern Recognition - one of 28 NVIDIA papers at the virtual event. It's also available on the AI Playground, where anyone can experience our research demos firsthand.

AI Steals the Show In a nod to classic heist movies (and a hit Netflix show), NVIDIA researchers put their talking-head GAN model through its paces for a virtual meeting. The demo highlights key features of Vid2Vid Cameo, including facial redirection, animated avatars and data compression.

These capabilities are coming soon to the NVIDIA Maxine SDK, which gives developers optimized pretrained models for video, audio and augmented reality effects in video conferencing and live streaming.

Developers can already adopt Maxine AI effects including intelligent noise removal, video upscaling and body pose estimation. The free-to-download SDK can also be paired with the NVIDIA Jarvis platform for conversational AI applications, including transcription and translation.

Hello from the AI Side Vid2Vid Cameo requires just two elements to create a realistic AI talking head for video conferencing: a single shot of the person's appearance and a video stream that dictates how that image should be animated.

Developed on NVIDIA DGX systems, the model was trained using a dataset of 180,000 high-quality talking head videos. The network learned to identify 20 key points that can be used to model facial motion without human annotations. The points encode the location of features including the eyes, mouth and nose.

It then extracts these key points from a reference image of the caller, which could be sent to other video conference participants ahead of time or re-used from previous meetings. This way, instead of sending bulky live video streams from one participant to the other, video conferencing platforms can simply send data on how the speaker's key facial points are moving.

On the receiver's side, the GAN model uses this information to synthesize a video that mimics the appearance of the reference image.

By compressing and sending just the head position and key points back and forth, instead of full video streams, this technique can reduce bandwidth needs for video conferences by 10x, providing a smoother user experience. The model can be adjusted to transmit a differing number of key points to adapt to different bandwidth environments without compromising visual quality.

The viewpoint of the resulting talking head video can also be freely adjusted to show the user from a side profile or straight on, as well as from lower or higher camera angles. This feature could also be applied by photo editors working with still images.

NVIDIA researchers found that Vid2Vid Cameo outperforms state-of-the-art models by producing more realistic and sharper results - whether the reference image and the video are from the same person, or when the AI is tasked with transferring movement from one person onto a reference image of another.

The latter feature can be used to apply the facial motions of a speaker to animate a digital avatar in a video conference, or even lend realistic expression and movement to a video game or cartoon character.

The paper behind Vid2Vid Cameo was authored by NVIDIA researchers Ting-Chun Wang, Arun Mallya and Ming-Yu Liu. The NVIDIA Research team consists of more than 200 scientists around the globe, focusing on areas such as AI, computer vision, self-driving cars, robotics and graphics.

Our thanks to actor Edan Moses, who performed the English voiceover of The Professor on La Casa De Papel/Money Heist on Netflix, for his contribution to the video above featuring our latest AI research.
LINK: https://blogs.nvidia.com/blog/2021/06/24/vid2vid-cameo-ai-research-vid...
See more stories from nvidia

Most recent headlines

05/01/2027

Worlds first 802.15.4ab-UWB chip verified by Calterah and Rohde & Schwarz to be demoed at CES 2026

Worlds first 802.15.4ab-UWB chip verified by Calterah and Rohde & Schwarz to be ...

20/11/2026

Hearst TV Names Shawnda Adams to Lead KOCO-TV Oklahoma City

Share {alert(Link copied)}).catch(err => {console.error(Failed to copy: , err)}) data-platform=copy_link data-action= class=font-ui-heading py-2 text-sm tex...

20/11/2026

FCC Modernizes Drone Communications, Detection Rules

Share {alert(Link copied)}).catch(err => {console.error(Failed to copy: , err)}) data-platform=copy_link data-action= class=font-ui-heading py-2 text-sm tex...

20/11/2026

OBSBOT Unveils 2 New PTZR Cameras

Share {alert(Link copied)}).catch(err => {console.error(Failed to copy: , err)}) data-platform=copy_link data-action= class=font-ui-heading py-2 text-sm tex...

20/11/2026

New DigitalGlue Mac App Identifies Media Production Hurdles

Share {alert(Link copied)}).catch(err => {console.error(Failed to copy: , err)}) data-platform=copy_link data-action= class=font-ui-heading py-2 text-sm tex...

20/11/2026

Beam Dynamics Introduces MCP for Faster, Smarter Technolo...

Beam Dynamics has introduced new Model Context Protocol (MCP) support, expanding how customers can work with the live operational data held in Beam. The new cap...

20/11/2026

Re:Vision Effects Cyber Week Sale Nov 24 to Nov 30

Re:Vision Effects Cyber Week Sale Nov 24 to Nov 30 Pierre Jasmin November 19, 2026 0 Comments Long time friends of the COW are up with a big sale! Us...

20/11/2026

Avid is offering up to 33% off annual subscriptions!

Avid is offering up to 33% off annual subscriptions! Brie Clayton November 19, 2026 0 Comments Hero image courtesy Deposit Photos Pro Tools Pro To...

19/11/2026

Live From the UFL Draft: League's First Live Draft Broadcast Is Set To Innovate, Bring Viewers Close to the Action

With a sports-bar vibe and all the teams onsite in a huge atrium in Ballpark Vil...

19/11/2026

SVG Students To Watch: Niamh Campbell, Seton Hall University

The senior from Phoenix and England is balancing life as a Division I tennis player with a growing career behind the microphone The future of live sports produ...

19/11/2026

RED Connect Adds Native Colorfront Color Management for Live RED Camera Feeds

RED Digital Cinema and Colorfront have partnered to integrate native Colorfront color management into RED Connect, letting productions apply Colorfront's co...

19/11/2026

SMPTE Publishes Engineering Report on Content Provenance and Authenticity

SMPTE has published an engineering report summarizing an industry survey by its Study Group on Content Provenance and Authenticity in Media. The findings point ...

19/11/2026

CTech and Skyline Communications Partner on Integrated SATCOM Operations for Defense

CTech and Skyline Communications have announced a partnership to deliver integra...

19/11/2026

EITB Deploys SMPTE ST 2110 Infrastructure with Lawo HOME mc DSP Processing

Euskal Irrati Telebista (EITB), the Basque Country's public media group, has completed Phase 1 of a five-studio renewal with the commissioning of a Lawo IP ...

19/11/2026

Esports Foundation Expands Road to Esports Nations Cup 2027 with New Qualifiers, Larger Rosters and National Rankings

The Esports Foundation (EF) has announced an expanded Road to ENC 2027 for the E...

19/11/2026

AWSN Holds First Women's Sports Awards Show, VEON: Women Who Win, in New York City

All Women's Sports Network (AWSN) held its first awards show, VEON: Women W...

19/11/2026

Gracenote Extends DIRECTV Partnership for Sports Hubs

Gracenote, the content intelligence business unit of Nielsen, has extended its relationship with DIRECTV. Gracenote will continue to supply visual imagery and p...

19/11/2026

AES Technical Committee Expands Loudness Website with Dialogue Intelligibility Content

The Audio Engineering Society (AES) Technical Committee on Broadcast and Online ...

19/11/2026

NDI 6.4 Adds Recording and Playback APIs, HEVC 4:4:4 and Wi-Fi Receiver Buffer

NDI has released NDI 6.4, which adds Recording and Playback APIs, support for HEVC 4:4:4 and a configurable receiver buffer for Wi-Fi. The update also includes ...

19/11/2026

Lemonlight Announces Hero, an AI Production Operating System for Video

Lemonlight, the video production company behind more than 50,000 brand videos, has announced Hero, an AI production operating system, along with a new AI agent ...

19/11/2026

FanConnect Brings JPEG XS Beyond the Control Room, Cuts In-Venue IPTV Latency 89% for University of Maryland Athletics

In a first-of-its-kind deployment at XFINITY Center, the digital signage/display...

19/11/2026

NBCU Intros Peacock Performance View for Premier League Coverage

For the first time on a US streaming service, real-time player identification, contextual stats, and visual insights appear right in the live stream, available ...

19/11/2026

Dome Productions Director of Engineering Mike Johnson To Retire at Years End

His career runs from the 2-in. videotape era to Canada's first HD and UHD live-production facilities...

19/11/2026

SVG GFX Forum Returns to NYC on Feb. 24 to Unite Graphics Tech and Creative Leaders

The popular sports-graphics event features a new Chairperson and new venue...

19/11/2026

Josephine Explores the Lingering Impact of Violence Through a Child's Eyes

Beth de Ara jo, Gemma Chan, Channing Tatum and Mason Reeves attend the premiere of Josephine at Eccles Theatre on January 23, 2026, in Park City, UT. (Photo b...

19/11/2026

Expressive E release Maestro Brass

New performance-focused brass library announced Expressive E's latest release brings together four meticulously sampled brass instruments: sousaphone, b...

19/11/2026

Korg introduce phase8 companion

Acoustic synthesizer gains free web editor The phase8 acoustic synthesizer has been an ongoing project for Korg Berlin. It was unveiled as an early prototyp...

19/11/2026

Impact GX49 & GX61 Mk3 from Nektar

MIDI controller keyboards upgraded Designed for users who want a full-size keyboard and hands-on DAW control but in a relatively compact footprint, Nektar&#...

19/11/2026

IK Multimedia unveil the iLoud Micro Monitor 2

Hugely popular compact monitor overhauled IK Multimedia's ultra-compact monitors have proven to be incredibly popular, and the company say that they'...

19/11/2026

Rohde & Schwarz brings company-wide technology and service expertise to critical infrastructure

Rohde & Schwarz brings company-wide technology and service expertise to critical...

19/11/2026

Global Impact, Local Action: Highlights from Nielsen Global Impact Day 2026

On October 8, 2026, Nielsen employees came together around the world to celebrate our fourteenth company-wide day of service, Nielsen Global Impact Day (NGID). ...

19/11/2026

Nexxen and Gracenote Partner to Expand Content-Based Targeting for CTV

Integrating Gracenote's content intelligence with Nexxen's unique CTV supply allows advertisers to more closely align campaigns with audience viewing ha...

19/11/2026

Nielsen Further Evolves Audio Measurement With Introduction of 1-Minute Qualifier

Enhancement Mirrors Local TV's Recent Move to 1-Minute Qualifier (1MQ) Upda...

19/11/2026

Samba Combines Streaming, Linear TV Data

Share {alert(Link copied)}).catch(err => {console.error(Failed to copy: , err)}) data-platform=copy_link data-action= class=font-ui-heading py-2 text-sm tex...

19/11/2026

Charting the Progress of Brazil TV 3.0 with VVC

Share {alert(Link copied)}).catch(err => {console.error(Failed to copy: , err)}) data-platform=copy_link data-action= class=font-ui-heading py-2 text-sm tex...

19/11/2026

Skydance to Combine Warner Bros. Television Group and Paramount Television Studios

Share {alert(Link copied)}).catch(err => {console.error(Failed to copy: , er...

19/11/2026

Australia's Seven Selects RTS Comms for New Headquarters

Share {alert(Link copied)}).catch(err => {console.error(Failed to copy: , err)}) data-platform=copy_link data-action= class=font-ui-heading py-2 text-sm tex...

19/11/2026

Richard Burki Expands the Possibilities of Immersive Audi...

As Founder and Creative Director of Future Phonic Studios, the Netherlands' first Dolby Atmos music studio, Richard Burki has spent more than a decade pushi...

19/11/2026

Boris FX Black Friday Sale - 25% Off Everything

Boris FX Black Friday Sale - 25% Off Everything Brie Clayton November 19, 2026 0 Comments Check out the Boris FX Black Friday sale on Monday, November...

19/11/2026

Virtual Production Finds Its Middle Ground with Blackmagic Design

Virtual Production Finds Its Middle Ground with Blackmagic Design Brie Clayton November 19, 2026 0 Comments How Neon Vision is using PYXIS 12K and Ult...

19/11/2026

FL Studio Announces Black Friday Discounts of Up to 50% Across Its Software Lineup

FL Studio Announces Black Friday Discounts of Up to 50% Across Its Software Line...

19/11/2026

TBS Drama Series Tagusari Brothers Uses DaVinci Resolve Studio

TBS Drama Series Tagusari Brothers Uses DaVinci Resolve Studio Brie Clayton November 19, 2026 0 Comments TBS ACT streamlines post production for broad...

19/11/2026

Beeble launches SwitchX 3.0 with dynamic environments and greater creative control

Beeble launches SwitchX 3.0 with dynamic environments and greater creative contr...

19/11/2026

NDI Releases Latest Version With Recording, Playback APIs

Share {alert(Link copied)}).catch(err => {console.error(Failed to copy: , err)}) data-platform=copy_link data-action= class=font-ui-heading py-2 text-sm tex...

19/11/2026

Newsmax Asks Court to Block Foxs Acquisition of Roku

Share {alert(Link copied)}).catch(err => {console.error(Failed to copy: , err)}) data-platform=copy_link data-action= class=font-ui-heading py-2 text-sm tex...

19/11/2026

HPA to Honor The Odyssey Director Christopher Nolan with Judges Award for Creativity and Innovation

Share {alert(Link copied)}).catch(err => {console.error(Failed to copy: , er...

19/11/2026

Australian Professional Leagues Renews Partnership with B...

Five-year deployment has automated APL's media supply chain from fixture to archive, managing over 200,000 assets Blue Lucy, the workflow orchestration and...

19/11/2026

Riding the Australian wave through co-pros

Riding the Australian wave through co-pros 11 November 2026 L-R: The Surfer, We Bury the Dead Australian stories are captivating global audiences and attracti...

19/11/2026

Screen Australia and Australian Film Television and Radio School announce the First Nations Producer Fellowship Program 2027

Screen Australia and Australian Film Television and Radio School announce the Fi...

19/11/2026

AES Adds Dialogue Intelligibility to Loudness Website

Share {alert(Link copied)}).catch(err => {console.error(Failed to copy: , err)}) data-platform=copy_link data-action= class=font-ui-heading py-2 text-sm tex...