
NVIDIA researchers are at the forefront of the rapidly advancing field of visual generative AI, developing new techniques to create and interpret images, videos and 3D environments.
More than 50 of these projects will be showcased at the Computer Vision and Pattern Recognition (CVPR) conference, taking place June 17-21 in Seattle. Two of the papers - one on the training dynamics of diffusion models and another on high-definition maps for autonomous vehicles - are finalists for CVPR's Best Paper Awards.
NVIDIA is also the winner of the CVPR Autonomous Grand Challenge's End-to-End Driving at Scale track - a significant milestone that demonstrates the company's use of generative AI for comprehensive self-driving models. The winning submission, which outperformed more than 450 entries worldwide, also received CVPR's Innovation Award.
NVIDIA's research at CVPR includes a text-to-image model that can be easily customized to depict a specific object or character, a new model for object pose estimation, a technique to edit neural radiance fields (NeRFs) and a visual language model that can understand memes. Additional papers introduce domain-specific innovations for industries including automotive, healthcare and robotics.
Collectively, the work introduces powerful AI models that could enable creators to more quickly bring their artistic visions to life, accelerate the training of autonomous robots for manufacturing, and support healthcare professionals by helping process radiology reports.
Artificial intelligence, and generative AI in particular, represents a pivotal technological advancement, said Jan Kautz, vice president of learning and perception research at NVIDIA. At CVPR, NVIDIA Research is sharing how we're pushing the boundaries of what's possible - from powerful image generation models that could supercharge professional creators to autonomous driving software that could help enable next-generation self-driving cars.
At CVPR, NVIDIA also announced NVIDIA Omniverse Cloud Sensor RTX, a set of microservices that enable physically accurate sensor simulation to accelerate the development of fully autonomous machines of every kind.
Forget Fine-Tuning: JeDi Simplifies Custom Image Generation Creators harnessing diffusion models, the most popular method for generating images based on text prompts, often have a specific character or object in mind - they may, for example, be developing a storyboard around an animated mouse or brainstorming an ad campaign for a specific toy.
Prior research has enabled these creators to personalize the output of diffusion models to focus on a specific subject using fine-tuning - where a user trains the model on a custom dataset - but the process can be time-consuming and inaccessible for general users.
JeDi, a paper by researchers from Johns Hopkins University, Toyota Technological Institute at Chicago and NVIDIA, proposes a new technique that allows users to easily personalize the output of a diffusion model within a couple of seconds using reference images. The team found that the model achieves state-of-the-art quality, significantly outperforming existing fine-tuning-based and fine-tuning-free methods.
JeDi can also be combined with retrieval-augmented generation, or RAG, to generate visuals specific to a database, such as a brand's product catalog.
https://blogs.nvidia.com/wp-content/uploads/2024/06/JeDi-cow-sculpture.mp4
New Foundation Model Perfects the Pose NVIDIA researchers at CVPR are also presenting FoundationPose, a foundation model for object pose estimation and tracking that can be instantly applied to new objects during inference, without the need for fine-tuning.
The model, which set a new record on a popular benchmark for object pose estimation, uses either a small set of reference images or a 3D representation of an object to understand its shape. It can then identify and track how that object moves and rotates in 3D across a video, even in poor lighting conditions or complex scenes with visual obstructions.
FoundationPose could be used in industrial applications to help autonomous robots identify and track the objects they interact with. It could also be used in augmented reality applications where an AI model is used to overlay visuals on a live scene.
NeRFDeformer Transforms 3D Scenes With a Single Snapshot A NeRF is an AI model that can render a 3D scene based on a series of 2D images taken from different positions in the environment. In fields like robotics, NeRFs can be used to generate immersive 3D renders of complex real-world scenes, such as a cluttered room or a construction site. However, to make any changes, developers would need to manually define how the scene has transformed - or remake the NeRF entirely.
Researchers from the University of Illinois Urbana-Champaign and NVIDIA have simplified the process with NeRFDeformer. The method, being presented at CVPR, can successfully transform an existing NeRF using a single RGB-D image, which is a combination of a normal photo and a depth map that captures how far each object in a scene is from the camera.
VILA Visual Language Model Gets the Picture A CVPR research collaboration between NVIDIA and the Massachusetts Institute of Technology is advancing the state of the art for vision language models, which are generative AI models that can process videos, images and text.
The group developed VILA, a family of open-source visual language models that outperforms prior neural networks on key benchmarks that test how well AI models answer questions about images. VILA's unique pretraining process unlocked new model capabilities, including enhanced world knowledge, stronger in-context learning and the ability to reason across multiple images.
VILA can understand memes and reason based on multiple images or video frames. The VILA model fa
North America Stories
31/03/2026
By Kiera Williams
Every year, a new group of emerging media and entertainment professionals steps into the HPA community - curious, talented, and ready to grow...
31/03/2026
MELBOURNE, Fla., March 31, 2026 - L3Harris Technologies (NYSE: LHX) will release its first quarter 2026 financial results before the market opens on Thursday, A...
31/03/2026
Photo courtesy of NASA. An illustration shows NASA's Perseverance rover operating on the surface of Mars....
31/03/2026
L3Harris Technologies and Comtech Telecommunications Corp. have collaborated to introduce the next evolution in resilient multi-orbit capability for today's...
31/03/2026
Rebel Sport leads ad investment as brands compete for a large, active and qualit...
31/03/2026
Share
Copy link
Facebook
X
Linkedin
Bluesky
Email...
31/03/2026
CueScript, a leading international developer of professional teleprompting solutions, is enhancing its platform with three innovations that elevate flexibility,...
31/03/2026
#GALSNGEAR will host a series of meetups, panel discussions, and networking events during the 2026 NAB Show, taking place April 18 22 in Las Vegas.
The organi...
31/03/2026
At the 2026 NAB Show (April 19-22, Las Vegas Convention Center, Booth #N1328), Imagine Communications is introducing AI assisted scheduling capabilities within ...
31/03/2026
High-density upgrade for mobile fleet and PSSI International Teleport gives PSSI more flexibility for premium live event coverage
PSSI and Appear today announc...
31/03/2026
Solid State Logic Enhances Virtualized System T Platform to Deliver Scalable Clo...
31/03/2026
OWC to Showcase Storage, Workflow Acceleration, and Reliability Solutions for Cr...
31/03/2026
Share
Copy link
Facebook
X
Linkedin
Bluesky
Email...
31/03/2026
OpenDrives, a leader in software-defined video and rich media storage management solutions, will demonstrate several new innovations at the 2026 NAB Show in Apr...
31/03/2026
At the 2026 NAB Show, SDVI will demonstrate how its Rally media supply chain management platform continues to give media operations teams the tools they need to...
31/03/2026
Boland Communications today announced that at the 2026 NAB Show, it will introduce its new QD4K315HDR10, a 31.5-inch QD-OLED monitor delivering exceptional colo...
31/03/2026
Dr. Paul Tapper, CEO of NUGEN Audio, will lead an essential industry discussion at the 2026 NAB Show, examining whether dialog intelligibility is poised to beco...
31/03/2026
BCNEXXT will use NAB 2026 to focus on how broadcasters can turn today's operational and market pressures into opportunity. Meeting with customers in the IAB...
31/03/2026
Mediagenix, a global leader in smart content solutions to profitably connect the right content to the right audience, will showcase its latest innovations at th...
31/03/2026
DHD announces that AI-based audio noise reduction is now available as a powerful new option for its XD3 IP Core. Developed in cooperation with ai-coustics, audi...
31/03/2026
PTZOptics will showcase its vision for intelligent video and advanced camera automation at NAB Show 2026 (#N1902). Through immersive demonstrations and technolo...
31/03/2026
Experience Commerce, a leading full-service digital marketing agency and part of the Cheil SWA Group, has secured the social media mandate for ICONIQA Hotels & ...
31/03/2026
Fabric, the entertainment industry's leader in data and operations solutions, will showcase a major expansion of its media technology platform at NAB 2026, ...
31/03/2026
Share
Copy link
Facebook
X
Linkedin
Bluesky
Email...
31/03/2026
Share
Copy link
Facebook
X
Linkedin
Bluesky
Email...
31/03/2026
Share
Copy link
Facebook
X
Linkedin
Bluesky
Email...
31/03/2026
Share
Copy link
Facebook
X
Linkedin
Bluesky
Email...
31/03/2026
Boris FX Acquires Vegas Pro, Sound Forge, and Acid Pro
Jessie Electa Petrov March 30, 2026
0 Comments
The acclaimed VFX developer bolsters its award-w...
31/03/2026
Berklee India Exchange Presents Anirudh Varma Collective The collective will present a workshop on Hindustani classical music and a concert in collaboration w...
31/03/2026
CERAWeek - dubbed the Davos of energy - is where policymakers, producers, techno...
30/03/2026
Manifold Technologies, a Germany-based provider of cloud infrastructure for live broadcast production, will demonstrate support for 400GbE COTS FPGA accelerator...
30/03/2026
Boland Communications will introduce its QD4K315HDR10, a 31.5-inch QD-OLED monitor, at NAB Show 2026 (Booth C3519, April 18-22). The company is also introducing...
30/03/2026
PTZOptics will demonstrate its Move 4K PTZ cameras and Horizon web-based control platform at NAB Show 2026 (Booth N1902).
Move 4K with Horizon is now available...
30/03/2026
Net Insight will demonstrate the next version of Nimbra Edge, its orchestration and control layer for live media services across multi-domain environments, at N...
30/03/2026
Appear ASA will exhibit at NAB Show 2026 (Booth W1531, April 19-22, Las Vegas). The company completed an IPO in November 2025.
Our customer-first approach is ...
30/03/2026
Harmonic has announced new capabilities for its sports streaming platform, covering multiview, programmatic advertising, in-stream advertising, and content wate...
30/03/2026
Ateme (Booth W1723) will demonstrate broadcast, streaming, and AI-driven media workflow solutions at NAB Show 2026.
GenAI and Agentic AI
Ateme will demonstrat...
30/03/2026
Bitmovin has announced new capabilities for Player Web X, its web video player, ...
30/03/2026
The 2026 NAB Show (April 18-22, exhibits April 19-22, Las Vegas Convention Center) will host Brazil's Minister of Communications, Frederico de Siqueira Filh...
30/03/2026
EVS will exhibit at NAB Show 2026 (Booth N1841), highlighting new products and updates across its live production portfolio, including the debut of T-Motion med...
30/03/2026
Solid State Logic will demonstrate its virtualized System T platform at NAB Show 2026 (Booth C6907). Demonstrations will include the VTE1 virtual DSP engine, ne...
30/03/2026
Globecast will exhibit at NAB Show 2026 (Booth W3335), highlighting its hybrid service model spanning satellite, IP, fiber, and cloud.
The company will demonst...
30/03/2026
The Alliance for IP Media Solutions (AIMS), Advanced Media Workflow Association (AMWA), and the Video Services Forum (VSF) have announced that the IP Showcase w...
30/03/2026
At NAB Show 2026 BBright will present a demonstration of its One Stream for the World concept, showing how a single ST 2110 playout stream can simultaneously ...
30/03/2026
OpenDrives will demonstrate new products at NAB Show 2026, with two locations in the West Hall: a pod (W3443-E) in the Sports Business Hub and a cabana at W1158...
30/03/2026
Behind The Mic provides a roundup of recent news regarding on-air talent, including new deals, departures, and assignments compiled from press releases and repo...
30/03/2026
The economics of live sports streaming have changed. New rights models, cloud production tools, and lower-cost distribution have made it possible for high schoo...
30/03/2026
CC-330 Husky. 2024 Eric Desbiens Photography.
Used with permission for the announcement and related communications. No residual rights....
30/03/2026
L3Harris Technologies will provide WESCAM CMX -8 sensor systems for integration on new Uncrewed Aircraft Systems from MDA Space, enhancing the Royal Canadian Na...
30/03/2026
Share
Copy link
Facebook
X
Linkedin
Bluesky
Email...