CVPR 2022 Logo

NVIDIA at CVPR 2022

Connect with NVIDIA researchers to learn more about the work presented at the Computer Vision and Pattern Recognition (CVPR) 2022 conference. Learn more about how NVIDIA Research collaborates with CVPR members to deliver AI breakthroughs across the community.

Webinar Replay: Building Custom Synthetic Data Generation Pipelines with NVIDIA Omniverse Replicator

NVIDIA Omniverse Replicator is an open and modular SDK that enables accurate 3D synthetic data generation (SDG) to accelerate training and performance of AI perception networks. Learn how Omniverse Replicator can help you produce physically accurate synthetic data at scale.

NVIDIA at CVPR 2022

Virtual Presentations

NVIDIA’s accepted papers at this year’s online CVPR feature a range of groundbreaking research in the field of computer vision. Papers ranging from Human Motion Forecasting to Extracting Triangular 3D Models, Materials, and Lighting From Images, explore the work NVIDIA is bringing to the CVPR community.

Improving Robustness to Tracking Errors with Affinity-Based Prediction

Xinshuo Weng, Boris Ivanovic, Kris Kitani, Marco Pavone | Coming Soon

ScePT: Scene-consistent, Policy-based Trajectory Predictions for Planning

Yuxiao Chen, Boris Ivanovic, Marco Pavone | Coming Soon

Motron: Multimodal Probabilistic Human Motion Forecasting

Tim Salzmann, Markus Ryll, Marco Pavone | Paper

IFOR: Iterative Flow Minimization for Robotic Object Rearrangement

Ankit Goyal, Arsalan Mousavian, Chris Paxton, Yu-Wei Chao, Brian Okorn, Jia Deng, Dieter Fox | Paper 

Neural Shape Mating: Self-Supervised Object Assembly with Adversarial Shape Priors

Yun-Chun Chen, Haoda Li, Dylan Turpin, Alec Jacobson, Animesh Garg | Coming Soon 

Modular Action Concept Grounding in Semantic Video Prediction

Wei Yu, Wenxin Chen, Songheng Yin, Steve Easterbrook, Animesh Garg | Paper 

X-Pool: Cross-Modal Language-Video Attention for Text-Video Retrieval

Satya Krishna Gorti, Noel Vouitsis, Junwei Ma, Keyvan Golestan, Maksims Volkovs, Animesh Garg, Guangwei Yu | Paper 

Unsupervised Pretraining with Region Priors for Object Detection

Amir Bar, Xin Wang, Vadim Kantorov, Colorado Reed, Roei Herzig, Gal Chechik, Anna Rohrbach, Trevor Darrell, Amir Globerson | Paper

Object-Region Video Transformers

Roei Herzig, Elad Ben-Avraham, Karttikeya Mangalam, Amir Bar, Gal Chechik, Anna Rohrbach, Trevor Darrell, Amir Globerson | Paper 

AUV-Net: Learning Aligned UV Maps for Texture Transfer and Synthesis

Zhiqin Chen, Kangxue Yin, Sanja Fidler | Paper

Neural Fields as Learnable Kernels for 3D Reconstruction

Francis Williams, Zan Gojcic, Sameh Khamis, Denis Zorin, Joan Bruna, Sanja Fidler, Or Litany | Paper

How Much More Data Do I Need? Estimating Requirements For Downstream Tasks

Rafid Mahmood, James Lucas, David Acuna, Daiqing Li, Jonah Philion, Jose M Alvarez, Zhiding Yu, Sanja Fidler, Marc T Law | Coming Soon 

BigDatasetGAN: Synthesizing ImageNet with Pixel-wise Annotations

Daiqing Li, Huan Ling, Seung Wook Kim, Karsten Kreis, Adela Barriuso, Sanja Fidler, Antonio Torralba | Paper

Generating Useful Accident-Prone Driving Scenarios via a Learned Traffic Prior

Davis Rempe, Jonah Philion, Leonidas Guibas, Sanja Fidler, Or Litany | Paper

Frame Averaging for Equivariant Shape Space Learning

Matan Atzmon, Koki Nagano, Sanja Fidler, Sameh Khamis, Yaron Lipman | Paper

Polymorphic-GAN: Generating Aligned Samples across Multiple Domains with Learned Morph Maps

Seung Wook Kim, Karsten Kreis, Daiqing Li, Antonio Torralba, Sanja Fidler | Coming Soon

Extracting Triangular 3D Models, Materials, and Lighting From Images

Jacob Munkberg, Jon Hasselgren, Tianchang Shen, Jun Gao, Wenzheng Chen, Alex Evans, Thomas Müller, Sanja Fidler | Paper

Panoptic SegFormer: Delving Deeper into Panoptic Segmentation with Transformers

Zhiqi Li, Wenhai Wang, Enze Xie, Zhiding Yu, Anima Anandkumar, Jose M. Alvarez, Tong Lu, Ping Luo | Paper

Not All Labels Are Equal: Rationalizing The Labeling Costs for Training Object Detection

Ismail Elezi, Zhiding Yu, Anima Anandkumar, Laura Leal-Taixé, Jose M. Alvarez | Paper

Bongard-HOI: Benchmarking Few-Shot Visual Reasoning for Human-Object Interactions

Huaizu Jiang, Xiaojian Ma, Weili Nie, Zhiding Yu, Yuke Zhu, Anima Anandkumar | Coming Soon

Non-parametric Depth Distribution Modeling based Depth Inference for Multi-view Stereo

Jiayu Yang, Jose M. Alvarez, Miaomiao Liu | Paper 

Searching the Deployable Convolution Neural Networks for GPUs

Linnan Wang, Chenhan Yu, Satish Salian, Slawomir Kierat, Szymon Migacz, Alex Fit Florea | Paper

Self-Supervised Pre-Training of Swin Transformers for 3D Medical Image Analysis

Yucheng Tang, Dong Yang, Wenqi Li, Holger Roth, Bennett Landman, Daguang Xu, Vishwesh Nath, Ali Hatamizadeh | Paper

Closing the Generalization Gap of Cross-silo Federated Medical Image Segmentation

An Xu, Wenqi Li, Pengfei Guo, Dong Yang, Holger Roth, Ali Hatamizadeh, Can Zhao, Daguang Xu, Heng Huang, Ziyue Xu | Paper

HyperSegNAS: Bridging One-Shot Neural Architecture Search with 3D Medical Image Segmentation using HyperNet

Cheng Peng, Andriy Myronenko, Ali Hatamizadeh, Vish Nath, Md Mahfuzur Rahman Siddiquee, Yufan He, Daguang Xu, Rama Chellappa, Dong Yang | Paper

GradViT: Gradient Inversion of Vision Transformers

Ali Hatamizadeh, Hongxu Yin, Holger Roth, Wenqi Li, Jan Kautz, Daguang Xu, Pavlo Molchanov | Paper

GroupViT: Semantic Segmentation Emerges from Text Supervision

Jiarui Xu, Shalini De Mello, Sifei Liu, Wonmin Byeon, Thomas Breuel, Jan Kautz, Xiaolong Wang | Paper

CoordGAN: Self-Supervised Dense Correspondences Emerge from GANs

Jiteng Mu, Sifei Liu, Shalini De Mello, Zhiding Yu, Nuno Vasconcelos, Xiaolong Wang, Jan Kautz | Paper

GLAMR: Global Occlusion-Aware Human Mesh Recovery with Dynamic Cameras

Ye Yuan, Umar Iqbal, Pavlo Molchanov, Kris Kitani, Jan Kautz | Paper

FreeSOLO: Learning to Segment Objects without Annotations

Xinlong Wang, Zhiding Yu, Shalini De Mello, Jan Kautz, Animashree Anandkumar, Chunhua Shen, Jose M. Alvarez | Paper

AdaViT: Adaptive Tokens for Efficient Vision Transformer

Hongxu Yin, Arash Vahdat, Jose M. Alvarez, Arun Mallya, Jan Kautz, Pavlo Molchanov | Coming Soon 

Efficient Geometry-aware 3D Generative Adversarial Networks

Eric Chan, Connor Zhizhen Lin, Matthew Chan, Koki Nagano, Boxiao Pan, Shalini De Mello, Orazio Gallo, Tero Karras, Sameh Khami, Jonathan Tremblay, Leonidas Guibas, Gordon Wetzstein | Paper

Watch It Move: Unsupervised Discovery of 3D Joints for Re-Posing of Articulated Objects

Atsuhiro Noguchi, Umar Iqbal, Jonathan Tremblay, Tatsuya Harada, Orazio Gallo | Paper 

When to Prune? A Policy towards Early Structural Pruning

Maying Shen, Pavlo Molchanov, Hongxu Yin, Jose M. Alvarez | Paper

Sound-Guided Semantic Image Manipulation

Seunghyun Lee, Wonseok Roh, Wonmin Byeon, Sang Ho Yoon, Chan Young Kim, Jinkyu Kim, Sangpil Kim | Paper

A Comprehensive Study of Image Classification Model Sensitivity to Foregrounds,Backgrounds, and Visual Attributes

Mazda Moayeri, Phillip Pope, Yogesh Balaji and Soheil Feizi | Paper

Attentive Fine-Grained Structured Sparsity for Image Restoration

Junghun Oh, Heewon Kim, Seungjun Nah, Cheeun Hong, Jonghyun Choi, Kyoung Mu Lee | Paper

Confidence Propagation Cluster: Unleash Full Potential of Object Detectors

Yichun Shen, Wanli Jiang, Zhen Xu, Rundong Li, Junghyun Kwon, Siyi Li | Paper

Blogs

Improving Diffusion Models as an Alternative To GANs

Deep Dive

NVIDIA Academic Hardware Grant Program

NVIDIA Academic Hardware Grant Program

The NVIDIA Hardware Grant Program helps advance artificial intelligence and data science by partnering with academic institutions around the world to enable researchers and educators with industry-leading hardware and software.

Applicants can request compute support from a large portfolio of NVIDIA products. Awardees of this highly-selective program will receive a hardware donation to use in their teaching or research.

Build Custom Synthetic Data Tools and Pipelines

Build Custom Synthetic Data Tools and Pipelines

NVIDIA Omniverse Replicator is an advanced framework built on Omniverse that enables physically accurate 3D synthetic data generation to accelerate the training and accuracy of perception networks. Omniverse Replicator is built on open standards like USD, PhysX, and MDL so it integrates seamlessly with your existing tools. Scale across multi-GPU and multi-node compute resources to batch massive synthetic data generation tasks.

Accelerate 3D Deep Learning Research

Accelerate 3D Deep Learning Research

NVIDIA’ Kaolin PyTorch library and NVIDIA Omniverse Kaolin App are a powerful suite of tools that simplify and accelerate 3D deep learning (DL) research. The Kaolin app leverages the NVIDIA Omniverse platform - built on USD and powered by NVIDIA RTX - to provide interactive tools that allow viewers to visualize 3D outputs of any DL model as it’s training, inspecting 3D data sets for inconsistencies, and more. 

Imaginaire

Imaginaire

Imaginaire is a PyTorch library that contains optimized implementations of several image and video synthesis methods developed at NVIDIA.

NVIDIA Applied Research Accelerator Program

NVIDIA Applied Research Accelerator Program

The NVIDIA Applied Research Accelerator Program supports research projects that have the potential to make a real-world impact through the deployment of NVIDIA-accelerated applications adopted by commercial and government organizations. This program accelerates development and adoption by providing access to technical guidance, hardware, and funding based on project requirements, maturity, and impact.

 

NVIDIA Maxine

Reinventing Real-Time Video Communications with AI

 

NVIDIA TAO

Create Custom Production-Ready AI Models without AI Expertise

 

NVIDIA Isaac Sim

Explore how to train and develop intelligent robots in simulation to be successfully deployed in the real world.

 

NVIDIA DRIVE Chauffeur and DRIVE

An AI-Powered Ride Through Silicon Valley

 

NVIDIA Clara Studio

Zero Coding Healthcare Solutions

 

Instant Nerf

Turn a collection of still images into a digital 3D scene in a matter of seconds with the implementation of neural radiance fields (NeRFs).

 

Sionna

Sionna is a GPU-accelerated open-source library for link-level simulations. It enables rapid prototyping of complex communication system architectures and provides native support for the integration of machine learning in 6G signal processing.

NVIDIA RESEARCH
AI PLAYGROUND

Discover our most recent AI research and the new capabilities deep learning brings to visual and audio applications. Explore the latest innovations and see how you can bring them into your own work. We'll update this page frequently with new demos and tools.

Resources

NVIDIA Developer Program

NVIDIA Developer Program

Get access to research grants, graduate fellowships, and technical resources to advance education and research by joining our Developer Program.

NVIDIA Deep Learning Institute

NVIDIA Deep Learning Institute

Learn and grow your technical knowledge by taking computer vision related self-paced courses and hands-on workshops through the Deep Learning Institute.

NVIDIA Inception

NVIDIA Inception for Startups

Inception provides 10,000 cutting-edge startups worldwide with critical go-to-market support, technical expertise, training, and introductions to funding opportunities to accelerate their business. See how program members in higher education, research and other industries are using computer vision to create innovative products

What Is NVIDIA Inception?

NVIDIA Inception is a free program designed to help your startup evolve faster through access to cutting-edge technology and NVIDIA experts, connections with venture capitalists, and co-marketing support to heighten your company’s visibility.

LIKE NO PLACE
YOU’VE EVER WORKED

Working at NVIDIA, you’ll solve some of the world’s hardest problems and discover never-before-seen ways to improve the quality of life for people everywhere. From healthcare to robots, self-driving cars to blockbuster movies, you’ll experience it all. Plus, there’s a growing list of new opportunities every single day. Explore all of our open roles, including internships and new college graduate positions.

Learn more about our current job openings, as well as university jobs.

Sign up to receive the latest news from NVIDIA.