This demo showcases the AMD Versal™ AI Edge Series Gen 2 VEK385 evaluation board’s ability to support streaming raw video data from 12 different CMOS image sensors concurrently and process 14 live streams out through the hardened ISP blocks in a single 2VE3858 adaptive SoC. It features a mix of sensors with different resolutions and […]
AMD Versal™ AI Edge Series Gen 2Hard ISP Featuring 12x Sensor Inputs Read More +
See how agentic AI workflows can use both local AMD hardware and cloud resources to improve privacy, performance, and cost efficiency. This demo showcases AI agents running with the AMD ROCm™ software platform. Private and data-sensitive tasks can remain on the local system, while more demanding workloads can be sent to cloud-based models when needed.
AI Agents on AMD – Secure Agent Computing at the Edge Read More +
Feisal Afzal discusses the growing market opportunities for MagikEye’s Invertible Light™ Technology across robotics, smart appliances, automotive systems, and emerging Physical AI platforms. The conversation explores how efficient depth perception can enable safer, smarter, and more capable machines operating at the edge. With increasing demand for spatial intelligence in the physical world, ILT is positioned
The Eyes of Physical AI: Market Opportunities for MagikEye’s ILT Technology Read More +
Jan Heller showcases ILT Pico, MagikEye’s compact developer platform built around its patented Invertible Light™ Technology. The system delivers precise depth perception with low power consumption and reduced compute requirements compared to conventional approaches. Designed for robotics, smart appliances, and edge AI applications, ILT Pico makes advanced machine vision more accessible to developers and innovators.
Introducing ILT Pico: Compact Depth Sensing for Physical AI Read More +
Watch a mobile robot powered by MagikEye’s Invertible Light™ Technology (ILT) navigate autonomously while detecting and avoiding obstacles in real time. This demonstration highlights how ILT delivers low-latency depth perception using a compact, power-efficient sensor architecture. The technology is designed for robotics, automation, and Physical AI applications where reliable spatial awareness is essential. Learn how
MagikEye ILT Enables Real-Time Robot Navigation and Obstacle Avoidance Read More +
Lattice Semiconductor demonstrates how the CertusPro-NX FPGA bridges an image sensor to a Raspberry Pi, performing on-device pre-processing and object detection before passing data to the host CPU. Sensor frames at 30 fps are fed into the FPGA, where an object detection model — trained on eight automotive object classes — runs locally and outputs
Smart Sensor Demo: On-Device Object Detection with Lattice CertusPro™-NX Read More +
Sébastien Taylor, VP of R & D at Au-Zone Technologies presents “From YOLO to SAM: Segmentation Models on Real Edge Hardware” at the May 2026 Embedded Vision Summit. Segmentation is fundamental to edge vision—from drivable surface detection to industrial inspection. But how do different approaches actually perform on resource-constrained hardware?… “From YOLO to SAM: Segmentation
Lattice Semiconductor demonstrates a complete end-to-end vision pipeline running entirely on an FPGA, from CSI camera input to screen output. Raw sensor data is processed by an on-chip ISP that converts it to a clean YUV 422 video stream in real time. The platform includes a dedicated ISP builder and tuning tool, allowing designers to
Lattice FPGA Embedded Vision: Real-Time ISP Pipeline with Integrated Tuning Tool Read More +
Tecphos demonstrates real-time video over Ethernet using a Lattice CrossLink-NX 40 FPGA at the Embedded Vision Summit. The FPGA captures raw image data from a MIPI image sensor, processes and scales it, and simultaneously delivers an uncompressed stream to the host while compressing it to JPEG on-chip. Compressed data is formatted into Ethernet packets and
Video over Ethernet on Lattice CrossLink™-NX FPGA: Microsecond Latency Demo Read More +
Mark Oliver, the VP of Marketing at Efinix demonstrates how multiple AI models can be compiled to run on dedicated AI accelerators implemented in the high performance Titanium FPGA family. He shows how Efinix supplied tools can be used to optimize an AI model to run on an AI accelerator delivering hardware level performance while
Here you’ll find a wealth of practical technical insights and expert advice to help you bring AI and visual intelligence into your products without flying blind.
Berkeley Design Technology, Inc.
PO Box #4446
Walnut Creek, CA 94596
