Seeed collaborates with partners to accelerate vision AI powered by NVIDIA Jetson and Metropolis

Last week, we were excited to exhibit at the Embedded Vision Summit with deployable computer vision and visual AI for product creators who want to bring visual intelligence to products. For Seeed’s hardware based on the NVIDIA Jetson edge AI and robotics platform, we showcased the following at the show:

We are excited about Computex, happening this week in Taipei and seeing how AI is transforming industries. In support of the NVIDIA Jetson platform, we collaborate with our edge AI partners Lumeo, CVEDIA, alwaysAI, STMicroelectronics, and Roboflow. We also look forward to more partners joining us to accelerate an end-to-end vision AI solution pipeline for all industries.

No-code Custom Video Analytics in Minutes

Seeed collaborates with Lumeo to offer a no-code custom intelligent video analytics solution, enabling any camera to run any AI model from basic to advanced analytics.

Easy-to-create AI applications with drag-and-drop builder in Lumeo

Vision AI in Retail

Waiting Time Analysis

In retail, you can use Lumeo to deploy occupancy monitoring, line crossing models, and more on Seeed’s reComputer J4012, powered by the NVIDIA Jetson Orin NX system on-module and the NVIDIA Metropolis platform. The mall’s integrator can quickly create and deploy a solution that uses existing cameras and VMS infrastructure to make the retail experience smoother. Check out Metropolis Spotlight: Lumeo Simplifies Vision AI Development.

Automate stock inventory

At the event, we also showcased Sparkcognition: automate ‘stock-out’ and inventory procedures for retail customers. SparkCognition’s award-winning AI solutions allow organizations to predict future outcomes, optimize processes, and prevent cyberattacks. SparkCognition partners with the world’s industry leaders to analyze, optimize, and learn from data, augment human intelligence, drive profitable growth, and achieve operational excellence.

Scalable Computer Vision Solutions at The Edge

Vision AI in Public Spaces

Seeed Studio collaborates with CVEDIA on intelligent security solutions running on Seeed’s Jetson-based edge AI modules with ready-to-use models for perimeter security, intrusion detection, crowd control, vehicle and people counting, vehicle and people classification, tripwire, zone analytics, etc. 

Computer Vision to Reduce Costs and Increase Efficiency

Vision AI in Manufacturing

Seeed Studio collaborates with alwaysAI for creating and deploying computer vision at the edge. In this use case, computer vision has been deployed in the Timberlab warehouse to pinpoint bottlenecks and patterns that help them save costs, bring operational efficiency, and drive revenues. Optimizing CNC operation, crane, and conveyor used to maximize throughput in the factory is crucial to bottom-line improvements.

Build and Deploy Computer Vision Models Faster with Roboflow

Roboflow empowers developers to build their own computer vision applications, no matter their skillset or experience. Roboflow provides all of the tools needed to convert raw images into a custom-trained computer vision model and deploy it for use in applications. Today, Roboflow supports object detection and classification models. Roboflow is optimized to get maximum performance from the NVIDIA Jetson line of edge AI modules by tailoring the drivers, libraries, and binaries specifically to its CPU and GPU architectures.

View the guide on deploying with Roboflow to NVIDIA Jetson. 

Steps to building, deploying, and improving a computer vision model with Roboflow

Accelerated inference via TensorRT on NVIDIA Jetson running JetPack 5.1.

Your JetPack SDK on the Jetson module will already have installed.

Image: roboflow/roboflow-inference-server-trt-jetson-5.1.1

docker run -it --rm -p 9001:9001 --runtime=nvidia roboflow/roboflow-inference-server-trt-jetson-5.1.1

NVIDIA TAO Toolkit and STM32Cube.AI

10x performance and vision AI energy savings

Seeed collaborates with STMicroelectronics on this demo featuring a Wio Lite AI based on STM32H7 running a real-time person-detection algorithm optimized using NVIDIA TAO Toolkit and STM32Cube.AI. The TAO-enabled model on the STM32 system determines whether people are present. If they are, it wakes up a downstream Jetson Orin module, enabling significant power savings. For such a system to be viable, the model running on the microcontroller must be fast enough to wake up the downstream device before the object leaves the frame. (Learn more at ST’s blog)

NVIDIA TAO Toolkit and Metropolis

World-Class Vision AI Accessible to Any Developer on Any Device

NVIDIA TAO Toolkit is a low-code AI framework that supercharges vision AI model development for any developer, in any service, on any device. TAO 5.0 is filled with new features, including vision transformer pretrained AI models, the ability to deploy models on any platform with standard ONNX export, automatic hyperparameter tuning with AutoML, and AI-assisted data annotation.

NVIDIA Metropolis puts the latest AI capabilities and research into the hands of developers through NVIDIA TAO Toolkit, Metropolis Microservices and the DeepStream SDK, as well as NVIDIA Isaac Sim for synthetic data generation and robotics simulation applications.

Metropolis makes it easier and more cost-effective for enterprises, governments, and integration partners to use world-class AI-enabled solutions to improve critical operational efficiency and safety problems. Seeed Studio works with NVIDIA Metropolis and its growing ecosystem of partners to help accelerate edge AI across industries.

Seeed and NVIDIA Jetson Ecosystem

Seeed is an Elite partner for edge AI in the NVIDIA Partner Network. Explore more carrier boards, full system devices, customization services, use cases, and developer tools on Seeed’s NVIDIA Jetson ecosystem page.

Join the forefront of AI innovation with us! Harness the power of cutting-edge hardware and technology to revolutionize the deployment of machine learning in the real world across industries. Be a part of our mission to provide developers and enterprises with the best ML solutions available.

About Author


June 2023