Discover How AI Is Transforming Video Analytics, at GTC

GPU Technology Conference to feature technical sessions for developers of AI-driven vision applications in smart cities, healthcare, industrial IoT, safety, robotics and more.
by Joanne Chang

AI-based video analytics is one of the fastest growing areas of computer vision. From creating AI models to building a real-time seamless video analytics pipeline, developers need powerful tools to simplify their workflows while achieving high performance and highly accurate results.

The GPU Technology Conference, Oct. 5-9, will feature NVIDIA technologies that help developers overcome all these challenges.

Running continuously across seven time zones over five days, GTC will showcase the latest breakthroughs in AI, intelligent video analytics, HPC, graphics and more to bring together thousands of developers, decision makers and researchers from around the world.

The virtual event will include several interactive live sessions in different time zones to deep dive into technical parts of NVIDIA’s vision AI tools. There will also be a huge catalog for vision AI sessions available on demand, and featuring the latest innovations and applications from NVIDIA and its partners across different industries — including smart cities, manufacturing, healthcare, workplace safety and telecommunications.

Here are some top technical sessions that will help developers understand NVIDIA tools to train, build and deploy AI-based video analytics applications.

Accelerating Vision AI Applications Using NVIDIA TAO Toolkit and Pre-Trained Models [A21333] 

Chintan Shah, senior product manager at NVIDIA, will walk through how to build high-performance vision AI models quicker and easier using NVIDIA TAO Toolkit and pre-trained models. The U.S. session is on Oct. 6 at 7 a.m. PT and the APAC session is on Oct. 8 at 3 p.m. KST.

Implementing Real-Time Vision AI Apps Using NVIDIA DeepStream SDK [A21337]

Amulya Vishwanath, product marketing lead at NVIDIA, will present the latest capabilities of the NVIDIA DeepStream SDK and demonstrate the reference apps using DeepStream SDK, TAO Toolkit and vision AI pre-trained models. The session is on Oct 6 at 9 a.m. PT.

Connect with the Experts: TAO Toolkit and DeepStream SDK for Vision AI/ Intelligent Video Analytics [CWEA001]

Schedule a one-on-one session with NVIDIA subject-matter experts to get your questions answered about NVIDIA DeepStream and TAO Toolkit. The session is on Oct. 6 from 2-3 p.m. PT. Seats are limited, so sign up today.

Here are some featured sessions that discuss how the NVIDIA Metropolis platform helps create value across a broad range of industries and use cases. For more sessions about AI-based video analytics, view the GTC session scheduler.

Building the AI City with NVIDIA Metropolis [A21335]

Adam Scraba, director of product marketing for Metropolis, will present how AI and the Metropolis intelligent video analytics platform is solving important problems across a range of industries. The session will cover the NVIDIA computing platform, solutions and ecosystem partners, along with specific examples of how AI is creating incredible value. The U.S. session is on Oct. 8 at 8 a.m. PT and the APAC session is on Oct. 7 at 10 a.m. CST.

Create, Deploy and Scale Intelligent Video Analytics with NVIDIA Metropolis [A22217]

Anthony Laskovski, developer relationship manager at NVIDIA, will present how the NVIDIA Metropolis platform helps businesses create, deploy and scale AI and IoT applications, with cutting-edge technology and an extensive developer ecosystem. Also, Michael Liou, vice president of Strategy and Growth at NVIDIA Inception startup, will share his company’s success story of deploying on Metropolis. The session is on Oct. 7 at 8 a.m. PT.

Driving Intelligence from Video at Scale Using AI in ArcGIS by Esri [A21370

David Yu, data scientist from Esri, will present how his company uses the NVIDIA DeepStream SDK to develop AI-powered video analytics solutions that help cities to save cost by understanding and visualizing city data.

Improve Patient Care with Everyday Sensors and Multimodal AI [A21257]

AI-enabled smart sensors cameras and microphones can act as eyes and ears that ensure public safety, improve patient care and enhance operational efficiency in healthcare facilities. Raghav Mani, product and developer relations manager at NVIDIA, will discuss how to use NVIDIA Clara Guardian to develop and deploy smart sensors with multimodal to monitor patients and improve their care. The session is on Oct. 6 at 12 p.m. PT

Contextual Intelligence and Anomaly Detection for Environment, Health and Safety [A21456]

Digitalization in the field of environment, health and safety enables organizations to go beyond traditional techniques to achieve the zero-incident objectives. Varghese Kollerathu, lead research engineer at Siemens, will present a visual understanding and perception-based solution using state-of-the-art AI technologies and NVIDIA DeepStream to ensure safety at workplaces.This session is on Oct. 6 at 8 a.m. PT.

Collaborative Traffic Cameras: Trajectory-Driven Approach Powered by NVIDIA Tools [A21708]

David Herman, CEO of DataFromSky, and Lenka Sediva, product manager at RCE Systems, will introduce the architecture of the trajectory-based technology behind DataFromSky and demonstrate practical examples and results with NVIDIA tools. This session is on Oct. 7 at 3 a.m. PT.

Edge AI Meets 5G: Efficient and Accelerated Intelligent Video Analytics [A22011] 

The nexus of 5G, IoT and MEC (Multi-access edge computing) is expanding the horizons of digital transformation. Ash Bhalgat, senior director of cloud & telco market development at NVIDIA, and Kuntal Chowdhury, senior vice president at Mavenir, will present an accelerated and efficient end-to-end 5G edge IVA solution running on the NVIDIA EGX platform and designed for public and private 5G customers.

NVIDIA Metropolis: GPU Computing in Modern Image and Video Processing Techniques [A22213]

Pooja Venkatesh, developer relations manager at NVIDIA, and Tinku Acharya, founder of Videonetics Technology, will cover the role of GPU computing in modern image and video processing techniques and showcase how NVIDIA Metropolis can be applied to various domains. This session will be on Oct. 8 at 7 a.m. IST.

NVIDIA Inception Premier Showcase: The Top AI Startups in North America [A22233]

Meet some of North America’s most advanced AI startups and see how they’re revolutionizing industries with the power of GPU computing, in the NVIDIA Inception Premier Showcase. Prasad Akella, CEO from NVIDIA Metropolis partner Drishti, will discuss how their video analytics and traceability solution helps solve critical manufacturing problems. Andrew Gostine, CEO at Whiteboard Coordinator, will discuss how AI makes hospitals smarter and safer with video analytics and NVIDIA GPUs.

Register for GTC

Early Bird pricing of $49 for a GTC Digital Conference pass is available through Friday, Sept. 25 — and is free if your email ends in .gov/.edu/.org/.mil. Further details are available on the GTC pricing page. Use our discount code CMAM for an additional 20 percent off.

Image credit: VIVACITY