Our solutions are production-ready software IPs specifically designed for speeding-up the adoption of deep learning, artificial intelligence and computer vision assets into OEM products and carefully optimized for the execution over embedded platforms at the edge, targeting the most relevant MPUs and AI accelerators on the market.
Some of the platforms we support:
- NXP iMX
- AMD Zynq Ultrascale+
- Raspberry Pi
- Nvidia Jetson
- Hailo-8 / Hailo-15
- Analog Devices Max78000
We are constantly expanding our supported platforms, so if your preferred board isn’t listed, reach out to us… we’ll work to include it!
STARTER KITS
ACCELERATE OEM DESIGNS
Struggling to bring AI-powered computer vision systems to market quickly? Our Starter Kits simplify the process, so you can focus on what matters: your product.
Deep Vision Starter Kits are built for OEMs looking to streamline the onboarding process of a modern AI-accelerated computer vision platform. The kits are based on Hailo-8™ AI accelerator and NXP i.MX 8M Plus MPU and include the Deep Vision Jumpstart Code, a source code repository (C++, GStreamer, and other scripts) with all the fundamental software elements of a vision project with accelerated AI.
- Quick prototyping
- Pre-configured AI-vision pipeline
- Run multiple deep learning models without loading the CPU
3D+AI
AFFORDABLE DENSE 3D RECONSTRUCTION
Need 3D vision but worried about the cost and complexity? Our AI-powered 3D reconstruction solution delivers dense, real-time depth maps on embedded devices, without breaking your budget.
To accelerate the integration of metric 3D vision for OEMs, Deep Vision has created 3D+AI, a software library for real-time dense 3D reconstruction from stereo cameras, using deep learning. This solution provides incredibly detailed 3D metric data in real time on embedded devices, making high-fidelity 3D vision accessible for various industrial applications.
- Metric depth perception and 3D reconstruction
- Produces dense point clouds and runs real time
- Optimized for embedded platforms supporting Hailo-8™ AI accelerator
DVI
AI ANOMALY DETECTION & QUALITY INSPECTION
Manual inspections slowing down your production? AI-powered defect detection automates and enhances quality control, improving efficiency and accuracy.
Defect Visual Inspection (DVI) is an advanced AI-driven visual anomaly detection solution designed for OEMs. Developed in partnership with Avnet Silica, DVI is available in two formats: Starter Kit → for easy and quick testing; or System on Module → for seamless integration into industrial production. Both options are based on NXP i.MX 8M Plus, ensuring high-performance embedded AI for real-time defect detection. Additionally, DVI is available as a standalone software library, adaptable to various computing platforms.
- AI-powered defect detection
- Available as hardware kits or standalone software library
- Training is done at the edge, no need for external PCs nor internet connection
TAGILY PRO
INDUSTRY-GRADE OCR FOR THE THOUGHEST CONDITIONS
Tired of OCR failures in non-ideal conditions? Our AI-driven OCR reads text and codes in any environment: blur, dirt, low light, reflects, or even partial occlusion.
Tagily Pro builds upon the success of Tagily, our OCR solution specialized for endurance sports, and extends it to industrial applications. This robust AI-powered OCR platform is designed to handle challenging conditions, such as: Low or saturated lighting; Motion blur and noise; Dirt, occlusions, and partial visibility..
Tagily Pro is fully adaptable to specialized industrial OCR needs, making it ideal for logistics, manufacturing, and automated inspections.
- Robust in extreme environments
- Highly adaptable to industry-specific needs
- Optimized for embedded systems and edge devices - no need to run on the cloud
