Existing cameras and NVRs
Confirm which camera streams are reachable, which codec and resolution are available, and whether a local network permits AI equipment to connect.
Products
AI video analytics products for existing CCTV systems.
Choose the form factor that matches your camera count, latency target, installation space, and integration requirements.
Existing CCTV AI upgrade
AI Edge Box for Video Analytics
2, 8, 16, and 32-channel edge AI video analytics box for upgrading existing RTSP and ONVIF CCTV systems with local detection, snapshots, event records, and platform integration.
- Retrofitting existing IP cameras, NVRs, and VMS platforms without replacing cameras
- Gas stations, factories, construction sites, parks, warehouses, campuses, and communities
- Projects that need local inference, event evidence, and structured alarm output
Front-end AI camera
Multi-Algorithm AI Camera for Local Video Analytics
8MP edge AI camera with an 8-core CPU and 3TOPS INT8 NPU for local multi-algorithm video analysis, alarm evidence, and custom model deployment.
- New or isolated camera points where local analysis is preferable to adding a separate edge box
- Warehouse entrances, equipment rooms, factory roads, fire lanes, loading areas, kitchens, parks, and unattended outdoor points
- Sites that need local configuration, event records, and daily recognition without external internet access
Smart recording and event search
AI NVR Smart Video Recorder
AI-powered NVR for local recording, false-alarm filtering, perimeter protection, intelligent search, event evidence, and ONVIF/RTSP camera compatibility.
- Projects where recording, event search, and AI alarm evidence should live in one appliance
- Parks, schools, communities, warehouses, factories, and small multi-camera sites
- Customers who need ONVIF/RTSP compatibility and local AI analysis without cloud dependency
Centralized analytics
AI Video Analytics Server
Centralize multi-channel AI video analytics for larger CCTV deployments and integration projects.
- Sites with many channels and a central equipment room
- VMS or platform integration projects
- Deployments requiring centralized model management and event routing
Custom algorithm production
XINHUOAI Custom Model Training Platform
Self-service custom AI model training platform for uploading field materials, running GPU training, evaluating models, and exporting deployable models to AI cameras, AI edge boxes, and AI NVRs.
- Customers whose target object, equipment state, workwear, sign, or behavior is not covered by a standard model
- Projects that need a repeatable data-training-evaluation-deployment workflow
- AI cameras, AI edge boxes, and AI NVR deployments that need new models after the first phase
Selection guide
Match product type to deployment scale.
A small gate or production line can start with a compact edge device. Multi-building sites often combine AI boxes, NVR management, and GPU servers for model density.
1-8 channelsEdge AI box or AI camera
8-64 channelsAI NVR or multiple edge boxes
64+ channelsGPU analytics server with VMS integration
Special sceneCustom AI model training and validation
Product selection notes
Choose the device around the camera estate and the operating workflow
The same detection model can be deployed in different places. An AI camera suits a dispersed or new point where the first analysis should happen at capture. An edge box adds analytics to reachable RTSP or ONVIF streams. An AI NVR combines recording and event management. A server supports a larger central channel mix. The project should start with available video access, network conditions, evidence needs, and the team that will act on the events.
Remote or new camera points
Where RTSP access, an NVR, or reliable broadband cannot be assumed, front-end AI cameras can analyse locally and send selected events over the available network.
Event and evidence workflow
Define what should be retained: snapshots, short video references, plate or object attributes, local records, external alarms, or all of these together.
Representative site test
Use actual camera footage to confirm image quality, rule settings, practical channel mix, and the results that an operator can review.
Project inquiry
Tell us your camera environment and required AI analytics.
For the first version this site uses a lightweight email-based inquiry flow. It avoids a database and keeps the English site simple until inquiry volume justifies a dedicated CRM integration.