Gidel at UVID 2026: FPGA Imaging, Sensor Fusion & Edge AI
Gidel will exhibit at UVID Dronetech 2026 on November 12 at Expo Tel Aviv, Pavilion 2, showcasing its FPGA-based imaging, Edge AI, and high-resolution image processing technologies, including FantoVision40-CXP12 and SkyBoost.
At the exhibition, Gidel will demonstrate FantoVision40-CXP12, combining real-time multi-sensor acquisition, deterministic FPGA image processing, Quality+ Compression, and NVIDIA Jetson AI. The demonstration will feature IR and RGB sensor fusion, supporting real-time object detection, tracking, and classification.
Gidel will also demonstrate SkyBoost, its FPGA-based accelerator for high-resolution RAW-to-JPEG image processing, significantly reducing processing time for demanding aerial imaging, defence, and other high-resolution applications.
Real-Time Multi-Sensor Imaging with FantoVision40-CXP12
FantoVision40-CXP12 combines high-bandwidth CoaXPress-12 image acquisition, FPGA processing, and NVIDIA Jetson AI in one compact embedded system.
The FPGA acquires and processes high-resolution image streams directly from the cameras, performing deterministic preprocessing, image enhancement, and compression before transferring the optimized image data to NVIDIA Jetson.
This heterogeneous architecture offloads image acquisition and supported processing operations from the CPU and GPU, leaving more Jetson computing resources available for sensor fusion and AI applications.
Key capabilities include:
- Multi-camera CoaXPress-12 acquisition with up to 4 × CXP-12 connections
- Real-time IR and RGB sensor fusion on NVIDIA Jetson
- Deterministic FPGA preprocessing with low latency
- Real-time image enhancement, including HDR processing
- Quality+ Compression to reduce downstream image bandwidth and storage requirements
- NVIDIA Jetson AI for object detection, tracking, and classification
- Multi-camera synchronization and processing
- Compact, low-SWaP architecture for embedded and mobile imaging systems
Live Demonstration: IR and RGB Sensor Fusion with Edge AI
At UVID 2026, Gidel will demonstrate FantoVision40-CXP12 simultaneously acquiring images from two CoaXPress-12 cameras, one infrared (IR) and one RGB Bayer, processing 13 MP images at 30 FPS.
The demonstration will combine deterministic FPGA processing, real-time image enhancement, Quality+ Compression, and NVIDIA Jetson AI.
The FPGA will process the incoming image streams and apply Quality+ Compression before transferring the image data to Jetson. This reduces downstream bandwidth requirements while preserving image quality and signal-to-noise ratio.
NVIDIA Jetson will perform real-time sensor fusion of the IR and RGB image streams, supporting AI-based object detection, tracking, and classification.
The demonstration will highlight how FPGA processing and NVIDIA Jetson AI work together to support high-bandwidth, low-latency, multi-sensor imaging applications.
SkyBoost: High-Resolution FPGA Post-Processing Acceleration
Gidel will also demonstrate SkyBoost, its FPGA-based image-processing accelerator for high-resolution RAW-to-JPEG workflows, including images exceeding 100 MP.
SkyBoost uses a PCIe FPGA accelerator installed in the host PC, with image processing performed directly on the FPGA. This offloads computationally intensive imaging operations from the host CPU or GPU and accelerates high-resolution post-processing.
At UVID 2026, Gidel will demonstrate SkyBoost accelerating high-resolution image processing from 30 seconds or more per image to more than two images per second. This performance can significantly shorten demanding imaging workflows involving large numbers of high-resolution images.
SkyBoost supports applications including aerial imaging, defence, intelligence, industrial inspection, scientific imaging, and other high-resolution vision systems.
For real-time image processing during acquisition, Gidel also develops SkyBoost-RT.
FPGA Imaging and Edge AI for UAV and Defence Applications
Modern UAV, counter-UAS, and defence imaging systems require high-resolution sensor acquisition, real-time processing, and AI capabilities within strict size, weight, and power constraints.
FantoVision integrates these functions into a compact embedded platform, making it suitable for applications including:
- UAV and airborne EO/IR imaging payloads
- Counter-UAS detection, tracking, and classification
- Intelligence, surveillance, and reconnaissance (ISR)
- Multi-sensor imaging and sensor fusion
- Autonomous imaging and navigation systems
- Border protection and perimeter surveillance
- Real-time image recording and streaming
- Embedded AI for defence and security applications
By combining deterministic FPGA image processing with NVIDIA Jetson AI, FantoVision supports high-performance imaging applications that require real-time operation close to the sensors.
See FantoVision and SkyBoost at UVID 2026
Visit Gidel at UVID Dronetech 2026 to experience FantoVision40-CXP12 with real-time IR/RGB sensor fusion and SkyBoost high-resolution FPGA image-processing acceleration.
Our team will be available to discuss FPGA-based imaging, high-bandwidth camera acquisition, real-time compression, and Edge AI integration for UAV, autonomous systems, and Aerospace & Defence applications.
See FantoVision in Action: Real-Time C-UAS Detection & Classification
Watch FantoVision demonstrate real-time C-UAS detection and classification, combining FPGA-based image processing with NVIDIA Jetson AI.
Download the FantoVision Datasheet
UVID 2026 Event Details
Date: November 12, 2026
Location: Expo Tel Aviv, Pavilion 2, Israel
Event: UVID Dronetech 2026
