Embedded Vision Summit
May 21–23 | Santa Clara, California
Meet Team Qualcomm at Embedded Vision Summit 2024
Embedded Vison Summit attracts a global audience of technology professionals from companies developing computer vision and edge AI-enabled products including embedded systems, cloud solutions and mobile applications. Visit us at booth 718 and experience our live demos and attend our live panel and sessions to learn more about our recent innovations such as Qualcomm AI Hub.
Session Overview
Wednesday May 22
OpenCV for High-Performance, Low-Power Vision Applications on Snapdragon
1:30 – 2:00pm
In this talk, we introduce a new element of Qualcomm’s computer vision tools suite: a version of OpenCV optimized for Snapdragon platforms, which allows developers to leverage and port their existing OpenCV-based applications seamlessly to Snapdragon platforms.
Speaker: Xin Zhong
Session Overviews
Thursday May 23
What’s Next in On-Device Generative AI
9:50 – 10:20am
In this talk, we will share Qualcomm’s vision of the compelling opportunities enabled by efficient generative AI at the edge. We will also identify the key challenges that the industry must overcome to realize the massive potential of these technologies. And we will highlight research and product development work that Qualcomm is doing to lead the way with an end-to-end system approach.
Speaker: Jilei Hou
Panel - Multimodal LLMs at the Edge: Are We There Yet?
10:20 – 11:10am
Mission City Ballroom B1-B5
Deploying Large Models on the Edge: Success Stories and Challenges
1:30 – 2:00pm
In this talk, Dr. Vinesh Sukumar will be talking about how Qualcomm has been successful in deploying large generative AI models for consumer and enterprise markets on the edge mapped to various use cases.
Speaker: Vinesh Sukumar
Session Overview
Tuesday May 21
Accelerating Model Deployment with Qualcomm AI Hub
1:00 - 4:00 pm
In this workshop we address the common challenges faced by developers migrating AI workloads from workstations to edge devices. Qualcomm aims to democratize AI at the edge, easing the transition to the edge by supporting familiar frameworks and data types. Additionally, we offer a range of tools to assist advanced developers in optimizing performance and power consumption for their AI applications.
In this session, we empower developers with knowledge and tools to efficiently deploy optimized models on real devices using the Qualcomm AI Hub. We’ll begin with an introduction to the Qualcomm AI Hub. We’ll then walk through examples showing how to optimize models using the AI Hub. Attendees will also have opportunities to speak with Qualcomm’s AI model deployment experts.
Join us for an immersive session where you’ll discover how to harness the power of AI Hub to streamline your deployment process and bring your AI applications to life in just minutes. This hands-on session is tailored for machine learning engineers and AI/ML application developers, including those working on Android, Windows, and IoT/AIoT applications. Bring your laptop (Windows, Mac or Linux) to fully engage with AI Hub. To participate in the hands-on experience, ensure you have Python (>=3.8, <=3.10) installed and the ability to run “pip install qai-hub”, or you can utilize Google Colab through your browser. We look forward to seeing you there!
Demo Overviews
DEMO 1: Qualcomm AI Hub
This demo is a comprehensive walkthrough of how to run an AI model from the AI Hub. You will get a chance to talk to the AI Hub team and see how AI Hub can help you streamline your CI/CD workflows.
Visit our Colab notebook and see how Qualcomm AI Hub makes it easy to deploy optimized models on real devices in minutes.
DEMO 2: AI PC with Snapdragon® X Elite
Experience the power of Llama2 on device with our locally running chatbot and how you can accelerate your AI workloads with the Snapdragon X Elite.
Demo 3: Qualcomm Linux and Qualcomm RB3 Gen 2 development kit (Private Preview)
Learn how you can run parallel inferencing for IoT use cases as we showcase four AI models running simultaneously on the RB3 Gen 2 development kit. Powered by Qualcomm Linux and bundled SDKs, this demo highlights the seamless integration and performance of multiple AI models, opening new possibilities for advanced applications.
DEMO 4: Large Multimodal Model (LMM) on Mobile devices
Demonstrating Large Language and Vision Assistant (LLaVA), a 7+ billion parameter large multimodal model (LMM) that can accept multiple types of data inputs, including text and images, and generate multi-turn conversations about an image. This LMM runs at a responsive token rate on device, which results in enhanced privacy, reliability, personalization, and cost.
Check out the tools and developer resources available in the Qualcomm AI Stack, and get started with building your AI-enabled apps.
