Explore featured sessions on computer vision solutions.

See the latest vision-AI advancements in developer tools, accelerated research, smart spaces, and deploying AI at the edge. With innovation happening across many industries, you won’t want to miss all the exciting use cases and discoveries that will be presented at this GTC.


Jensen Huang | NVIDIA | Founder and CEO

Watch the keynote replay to hear Jensen Huang's insight into how NVIDIA is driving the rapid pace of technology advancements to help solve the world's toughest challenges.

Featured Speakers

Xin Wang
John Madsen

Bridging the worlds of NVIDIA Metropolis / EGX and Video Management Systems

John Madsen
Distinguished Research Engineer, Milestone Systems

Juan Rodolfo Alvarez Padilla

Intelligent Video Analytics: The brains of the business

Juan Rodolfo Alvarez Padilla  
Head of CV, Gesta Labs

Tae-Ho Kim

Sessions By Topic

The Latest in Intelligent Video Analytics

  • Accelerating the Development of Next-Generation AI applications with DeepStream 6.0

    • Alvin Clark, Product Marketing Manager for DeepStream, NVIDIA
    • Carlos Garcia-Sierra, DeepStream Product Manager, NVIDIA

    Edge AI and distributed processing applications are at the forefront of the deep learning revolution. Learn how DeepStream 6.0 can help you create the next big thing for retail, manufacturing, healthcare, smart cities, and beyond. 

  • How Cities, Stadiums, and Factories are Leveraging Edge AI and Metropolis

    • Ekaterina Sirazitdinova, Data Scientist, NVIDIA
    • Michael Israel, Chief Information Officer, The Kraft Group
    • James Alberque, GIS and Emerging Technology Manager, City of Raleigh, North Carolina
    • Derek Vote, Meat Scientist, JBS USA

    What do these companies have in common? The operation of their most valuable spaces and infrastructure is being streamlined, and management costs are being reduced. In this panel, experts will explore how AI is impacting their work, how edge computing is being deployed, and best practices to guide other smart spaces on their journey to more efficient management.

  • Bridging the worlds of NVIDIA Metropolis / EGX and Video Management Systems

    • John Madsen, Distinguished Research Engineer, Milestone Systems

    In this session, we’ll propose a new API specification that aims to standardize how Intelligent Video Analytics (IVA) apps integrate with video management systems. We’ll present a concrete implementation of a gateway that bridges IVA with the Milestone XProtect VMS.

  • Step-by-Step Guide to Starting, Operating, and Optimizing Vision AI Applications

    • Tae-Ho Kim, CTO, Nota 

    Learn how Nota tools, such as Nota AutoML’s NetsPresso model compression platform, empowers your AI journey on NVIDIA Jetson(s). Nota AutoML allows you to select the inference device, define optimal accuracy and latency setting, and generate the hardware-aware and cost-efficient model efficiently to lower the barrier to deploy models in production.

  • How to Quickly Develop and Stand Up Smart Infrastructure Solutions with AI LaunchPad and Metropolis

    • Anthony Laskovski, Developer Relations Manager, NVIDIA
    • Kevin Jones, Principal Product Manager, NVIDIA

    Combining cloud-native development, fully managed edge AI deployments, and instant access to EGX cloud instances lets you better focus on application development. Learn how to develop Metropolis applications that can be deployed using Fleet Command out to a globally distributed AI Infrastructure provided by AI LaunchPad.

Vision-AI across Industries

  • Combine Multi-Camera Tracking and Footfall to Reveal the Mall Part of the Multichannel Customer Journey

    • Nicolas Bouvattier, CEO, Digeiz

    Shopping malls are under pressure with commercial rents increasing, etailers’ declining profitability, environmental issues, excessive offer (USA), retail bashing, and stock prices (REITs). Learn how multi-camera tracking addresses these retail challenges. 

  • Building Safer Public Transportation with AI-Based Video Analytics

    • Johan Barthelemy, Lecturer, University of Wollongong, AU

    Enhancing the safety and operational efficiency of the world’s public transit is a worldwide imperative. The largest transportation network in Australia, Transport for New South Wales (TfNSW), is building an AI-enabled public safety system in collaboration with University of Wollongong (UOW).

  • Software 2.0 for Industry 4.0

    • Chitra Singh, Member of Technical Staff, Drishti

    Drishti is committed  to measuring, informing, and guiding the work at assembly lines of leading automotive, medical, and industrial manufacturing facilities. Learn how to perform intelligent video analytics at scale with cutting-edge computer vision and deep learning technologies, using only video data. 

  • The Rise of Smart Infrastructure: Automating Smart Spaces with NVIDIA Metropolis and Edge AI

    • Adam Scraba, Director of Product Marketing, NVIDIA
    • Debraj Sinha, Product Marketing Manager, NVIDIA

    From retail shops to warehouses to our city streets, every space is becoming smarter because of AI-enabled vision applications. In this talk, NVIDIA will share how the world’s smartest spaces are using the NVIDIA Metropolis AI application framework to accelerate time to impact—from development to deployment and management. 

  • Intelligent Video Analytics: The brains of the business

    • Margaret Amori, US Public Sector Inception Manager, NVIDIA
    • Mauricio Mesones, CEO, Jebi AI
    • Alex Rozgo, Lead Simulation and AI Engineer, Vertex Studio
    • Juan Rodolfo Alvarez Padilla, Head of CV, Gesta Labs
    • Roberto Fernandino, CEO, SVA Tech

    Join this panel of Inception startups in the LATAM region to discuss how Intelligent Video Analytics will accelerate business transformation and why forward-looking organizations that adopting the technology will be uniquely positioned to reap its limitless benefits.

Computer Vision - Research

  • Perceive, Reason, Act: Closing the AI Loop

    • Gal Chechik, Director of AI, NVIDIA

    AI can help build systems that interact with their environment, with people, and with other agents in the real world. This poses algorithmic challenges. We’ll go over research into these challenges, focusing on modelling the high-level structure of a visual scene; using compositional structures in attribute space to learn from descriptions without any visual samples; and teaching agents new concepts without labels by using elimination to reason about their environment.

  • 3D Perception for Semantic Scene Understanding

    • Angela Dai, Professor, Technical University of Munich

    Remarkable progress has been made in recent years in 2D visual understanding leveraging deep neural networks. But they largely make predictions in the 2D domain, rather than the underlying 3D structure of the world around us. We propose to leverage geometric structural priors for 3D object perception from 2D images. We'll demonstrate that learned implicit 3D priors (e.g., view-invariance) can be used to benefit 2D semantic scene-understanding tasks.

  • 3D Reconstruction for Game Development

    • Bo Yang, Manager of Graphics and Vision Team, Tencent

    Learn about 3D object reconstruction methods for game development. We'll discuss differentiable rendering techniques for mesh and implicit surface, so a 3D object could be refined iteratively by comparing its projections with 2D images, 3D game character creation based on a few 2D images, and how we use a deep learning approach for implicit surface estimation.

  • Generalized Neural Implicits, and Humans Interacting in the 3D World

    • Gerard Pons-Moll, Professor, University of Tübingen

    The field of 3D shape representation learning and reconstruction has been revolutionized by combinations of neural networks with implicit and field representations. Discover a new class of neural implicit models that are robust, preserve detail, can be trained from raw 3D scans, and are controllable by the user.

  • Computer Vision Research at NVIDIA

    • Jan Kautz Vice President, Learning and Perception Research NVIDIA

    In this talk, learn about the latest NVIDIA-powered advances in computer vision.

Computer Vision - Image Processing

  • How AI is Helping to Reduce Risk and Enable Text-Intensive Automation

    • Oscar Guerra, Inception Program Manager, NVIDIA
    • Armin Bauer, Co-Founder and Managing Director Technology, IDnow
    • Rachel Kirkham, Executive Director, Audit Risk and Analytics, MindBridge
    • Shay Strong, VP of Analytics, ICEYE
    • Susana Latorre, Business Development Manager, Munich RE
    • Filip Graliński, Chief Data Scientist, Applica

    The adoption of AI is wide and deep, and few industries exemplify this diversity better than finance. Learn how our startups are using computer vision to make financial services more efficient, flexible, and profitable.

  • Cryo-RALIB: A Modular Library for Accelerating Alignment in Cryo-EM

    • Szu-Chi Chung, Assistant Professor, Department of Applied Mathematics at National Sun Yat-sen University

    Cryo-EM has become a rapid structure determination method that permits capture of dynamical structures of molecules in solution, as was recently demonstrated by determining the COVID-19 virus's spike protein in March 2020. We introduce Cryo-RALib, a library that expands the functionality of CUDA library used by GPU ISAC, and we connect the cryo-EM image analysis with the Python data science stack to make it easier for users to perform data analysis and visualization.

  • AI Impacts to Healthcare and Life: A Journey of Infusing and Democratizing AI to Healthcare by VinBrain

    • Steven Truong, CEO, VinBrain Joint Stock Company

    VinBrain's team will share their cutting-edge work/DrAid# (winner of 2021 ACM SIGAI Industry Award for Excellence in Artificial Intelligence), covering medical imaging use cases developed through multiple medical modalities and big data, including CXR, CT scan, MRI, etc. See how NVIDIA’s pre-trained models, Clara SDKs, Transfer Learning Toolkit, Triton, and GPUs are integrated.

  • Progressive Semantic Segmentation

    • Minh Hoai, Head of Applied Perception Group, VinAI Research

    We present MagNet, a method to segment high-resolution images without overloading GPU memory usage or losing the fine details in the output segmentation map.

  • The Future of Identity in a Post-COVID World

    • Kedar Kulkarni, CEO, HyperVerge

    Nearly $56 billion has been lost by consumers, and nearly $721 billion is expected to be lost as a result of identity fraud in 2021. COVID has further accelerated the number of cases of account takeover, identity theft, and fraudulent transactions in a remote-first world. We’ll discuss the best methods of fraud prevention, and how organizations around the world are leveraging fraud data. 

  • Addressing Generalization and Scalability Challenges in Satellite Imagery Analysis Using NVIDIA GPUs and Deep Learning

    • Philipe Ambrozio Dias, Research Associate, Oak Ridge National Laboratory (ORNL) 
    • Hsiuhan Lexie Yang, Research Scientist, Oak Ridge National Laboratory (ORNL)

    Applying deep learning models for Earth observation is hampered by major challenges. We’ll share our experiences on extracting building footprint and roads from satellite imagery datasets and using multi-GPU and multi-node HPC platforms to use NVIDIA DGX machines and ORNL’s Summit supercomputer. We’ll also discuss our research using Jetson edge-computing devices and few-shot learning for unmanned aerial survey utility pole inspection and damage assessment.

Explore More Conference Topics

Explore All Session Topics

NVIDIA Developer Program

Get the advanced tools and training you need to successfully build applications on all NVIDIA technology platforms.

Accelerate your Startup

Explore the startup track at GTC to learn how NVIDIA Inception can fuel your growth through go-to-market support, world-class training, and technology assistance.

Get Hands-On Training

Interested in developing key skills in AI, accelerated data science, or accelerated computing? Get hands-on instructor-led training from the NVIDIA Deep Learning Institute (DLI) and earn a certificate demonstrating subject matter competency.