Building GPU-Accelerated Streaming AI Pipelines With Holoscan SDK

Join us to deep dive on how to use Holoscan SDK to deploy an AI model conveniently to achieve an end-to-end GPU-accelerated streaming AI application. The NVIDIA Holoscan SDK is part of NVIDIA Holoscan, the AI sensor processing platform that combines hardware systems for low-latency sensor and network connectivity. With the sensor-processing capabilities for the next generation of AI applications at the edge, Holoscan SDK provides ready-to-use GPU-accelerated operators and an easy way to build streaming AI pipelines for a variety of domains, either in C++ or Python. We'll describe the core concepts of Hololscan SDK, and take Yolo Model as an example to show how to deploy an object detection application within Holoscan SDK. Besides that, we'll show how to interoperate with popular preprocessing libraries (CV-CUDA, cuCIM and OpenCV) in Holoscan SDK to create customized operators within the existing pipeline.