SIJUNG awakens spatial intelligence
with a Vision Station that thinks and acts on its own.
We give imaging devices an independent reasoning capability.
Not merely capturing footage — but truly understanding and deciding
at the edge, in real time.
The camera becomes eyes.
And then a brain.
NPU-based standalone vision computing hardware. An edge device that runs full AI inference without a network. Directly compatible with existing CCTV infrastructure.
🔌 Compatible with Existing InfrastructureA dedicated operating system optimized for vision computing. AI models become apps, running independently on the runtime. The ecosystem expands through the SDK.
🧩 Open SDK ArchitectureAn on-device AI inference engine that doesn't rely on the cloud. Processes video in real-time with Zero-copy Pipeline, operating fully autonomously even in air-gapped environments.
⚡ Sub-50ms InferenceVideo is transmitted to central servers for processing. System failure when the network goes down. Data leakage risk. High operational costs.
Video never leaves the site. Processed instantly on-site, operating autonomously regardless of network status.
Beyond simple security — everywhere a nation's sensory nervous system is needed.
When cameras gain a brain,
infrastructure begins to make its own decisions.
That is the future SIJUNG is designing.
SIJUNG Inc. is a research spinoff established through technology investment from the Korea Research Institute of Standards and Science (KRISS).
Since our founding in 2016, we have combined computer vision and AI
to solve problems across aviation, maritime, transportation, and beyond.
Through collaboration with the Korea Research Institute of Standards and Science, we developed visibility measurement technology and founded a company integrating Vision AI on that foundation. By combining traditional optical technology with AI, we deliver more precise and efficient visibility solutions — and have since expanded into privacy protection as well.
SIJUNG solutions installed and operating at public institutions across Korea and overseas
State-of-the-art deep learning research delivering high accuracy and stability. Industry-specific models solving diverse business challenges.
Advanced vision AI including object detection, tracking, and classification through video data analysis. Real-time processing combined with privacy protection.
Clearly explains AI model reasoning to users. Strengthens trust in AI and meets regulatory and ethical requirements.
Click an image to enlarge
Secure on-premise AI solutions deployable even in air-gapped environments