On-premise AI video analytics — private & GDPR-compliant
Camlytics is an AI video analytics platform where all video processing happens on your own hardware. No video, no images, and no camera footage ever leave your facility — the AI engine runs locally and sends only anonymous numeric events (counts, occupancy values, alerts) to the Camlytics Service cloud portal. This makes Camlytics the right choice for organizations that need to protect sensitive footage while still getting a full-featured analytics dashboard, API, and webhooks.
How on-premise video analytics works
Camlytics installs as a Windows application on a server or PC connected to your camera network. The AI engine receives RTSP streams from your IP cameras or NVR and processes them in real time using computer vision models that run entirely on your local hardware.
The results — counting events, zone occupancy values, motion alerts — are forwarded to the Camlytics Service cloud portal, which provides the analytics dashboard, REST API, webhooks, and reporting. Video itself is never part of this transfer: only small, anonymous event records (e.g., "vehicle entered zone B at 09:14") reach the cloud.
This architecture gives you the privacy of on-premise processing combined with the convenience of a cloud-hosted interface accessible from any browser.
GDPR & privacy compliance
Cloud-based video analytics platforms create significant GDPR exposure: video containing identifiable individuals is transmitted to and stored on third-party servers, often in unknown jurisdictions.
With Camlytics, no biometric or personal data leaves your premises. The software outputs only anonymous aggregate counts and events (e.g., "3 people entered zone A at 14:32") — never images or video clips. This architecture is inherently compliant with GDPR Article 25 (data protection by design) and avoids the need for a Data Processing Agreement with a video-hosting vendor.
On-premise video processing vs cloud video analytics
- Privacy: Camlytics keeps all video local; traditional cloud solutions upload footage to external servers
- Processing: AI runs on your hardware — no video bandwidth consumed, no cloud compute cost
- Dashboard & API: Events (not video) are sent to the Camlytics Service cloud portal for reporting, API access, and webhooks
- Cost: Per-channel subscription with no video-storage or video-bandwidth fees
- Control: You own the camera footage; only anonymous event data passes through the cloud portal
Hardware requirements
Camlytics runs on standard Windows servers — no GPU is required for most use cases. A modern multi-core CPU is sufficient for processing 4–8 camera streams simultaneously. GPU acceleration is supported for larger deployments.
There are no proprietary hardware locks or appliance requirements. You can run Camlytics on existing on-site servers, a dedicated mini-PC, or an industrial edge device.
See the full hardware requirements PDF for channel capacity guidelines by CPU tier.
Industries that benefit most
- Healthcare — patient flow analytics in hospitals and clinics without HIPAA-sensitive video leaving the building
- Government & public sector — facilities with strict data-sovereignty requirements where camera footage must not leave the premises
- Finance & banking — branch foot-traffic analysis compliant with internal data-security policies
- Retail & hospitality — customer flow and occupancy insights without sharing footage with cloud vendors
- Education — campus occupancy monitoring compliant with FERPA and local data-protection laws
Analytics capabilities
On-premise video processing does not limit what Camlytics Service can detect and measure:
- People counting — bidirectional line crossing, zone occupancy
- Vehicle counting — cars, trucks, buses, bikes
- Occupancy monitoring — real-time space utilization
- Demographics — age and gender estimation
- Queue management — queue length and wait time
- Perimeter security — intrusion and motion alerts
Pricing Downloads Dashboard demo Hardware requirements
Related solutions
Explore what you can measure with Camlytics while keeping all video on your premises:
- People counting solution — count visitors with 96% accuracy using any camera
- Occupancy monitoring — track live space occupancy with video processed on-site
- Parking analytics — monitor parking lot occupancy with local video processing
- Video analytics API — access analytics events via the Camlytics Service REST API
Frequently Asked Questions
No. Camlytics Service processes all video locally on your own hardware. Only anonymous numeric events (such as counts and zone occupancy values) are transmitted to cloud dashboard. No video frames, images, or identifying data ever leave your facility.
Yes. Because no personal data or video is transmitted to external servers, Camlytics is inherently aligned with GDPR data-protection-by-design principles (Article 25). The software outputs only anonymous aggregate counts, not biometric or identifiable information. You remain the sole data controller with no third-party data processor handling your footage.
The legacy Camlytics Single (offline) edition can run without any internet connection — analytics, local event storage, and file-based exports all work fully air-gapped. However, Camlytics Service — the current main product — requires an active internet connection. Event viewing, the cloud dashboard, REST API, and webhooks are all cloud-hosted features and are not available without connectivity.
Camlytics Service runs on standard Windows servers or PCs — no proprietary hardware or GPU is required for most deployments. A modern multi-core CPU (e.g., Intel Core i5/i7 or Xeon) is sufficient for 4–8 camera streams. Refer to the hardware requirements PDF on the downloads page for detailed capacity guidelines.
Cloud-based analytics send your camera footage to external servers for processing, creating privacy, latency, and cost risks. Camlytics Service processes everything locally: video stays on your network, there are no bandwidth costs, processing continues during internet outages, and you have full control over data retention and access.