Cameras + NVR
Existing RTSP cameras and supported Hikvision NVR channels connect inside each site.
RTSP ingestionVIDEO INFRASTRUCTURE / VIGILSYN
Every site.
Every camera.
One clear view.
From the cameras on your network to the people who need to see them. A self-hosted platform for live video, recording and camera control.
Your hardware. Your infrastructure. Your view.
THE CONNECTED VIDEO NETWORKEDGE → CLOUD
Local capture. Connected operations.V / 01
Reference architecture · deployment illustration
BUILT TO CONNECT
RTSP cameras / Hikvision NVRs / WebRTC & HLS / Tunnel
THE SIGNAL PATH
Local by design.
Connected by choice.
Keep camera traffic and control close to the source. Bring the views you choose into a shared Cloud, then into the applications your team uses.
Edge node
Local recording, camera discovery and PTZ control. Publish selected cameras to Cloud.
Tunnel uplink →Cloud node
Organize sites and cameras, authorize access and serve live streams to your viewers.
WebRTC / HLS delivery →Viewer + your app
Use the operator Viewer or integrate video with backend-issued, short-lived playback tokens.
Scoped playback accessCamera control and recording API requests can stay local to Edge. Viewer and application access use reachable Cloud endpoints; they do not require their own tunnel.
01 / THE PLATFORM
A local view.
A connected operation.
Cameras belong on your network. The right view should be available wherever your work takes you.
Cameras & NVROn your LAN
RTSP
EdgeCapture & control
Tunnel
CloudSites & delivery
Scoped tokens
Your applicationYour interface
THE EDGE NODE
The work starts where the cameras are.
Connect RTSP cameras or import channels from a Hikvision NVR. Keep recording, presets and camera control close to the source, with a local operator interface.
- Camera and NVR discovery
- Recording with time-based retrieval
- PTZ presets and API control
MADE FOR THE WORK BETWEEN INCIDENTS
Find the view.
Follow the moment.
The everyday work matters: knowing which camera you’re looking at, finding a recording, and giving the right person the right access.
Talk through your workflowLive, with context.
Organize cameras by site, region and group. Move from an overview to the camera you need, with WebRTC and HLS playback options.
Recordings you can find.
Browse camera recordings by time. Use named clips and configurable pre-roll and post-roll when your workflow needs a specific moment.
Control where it belongs.
Use supported camera or Hikvision NVR PTZ endpoints. List and use presets from an operator interface or your own application.
Access with boundaries.
Give applications scoped credentials and short-lived playback tokens. Manage Cloud and Edge capacity with online or offline licenses.
02 / YOUR INFRASTRUCTURE
Fit the system
to the site.
Start with your network, hardware and operating needs. Deploy the pieces you need, on servers you control.
A / LOCAL
At the edge.
Keep cameras, recording and local control on the site LAN. Deploy the Edge node beside your existing camera network.
Local operations / Existing hardwareB / CONNECTED
Across sites.
Connect Edge nodes to Cloud through a tunnel. Organize camera access and deliver streams without exposing every camera individually.
Multi-site / Central visibilityC / INTEGRATED
Inside your product.
Make live video part of an existing web application. Your backend handles authorization; your users stay in your experience.
Developer API / Signed playbackDocker deployment on Ubuntu. Offline image packages for restricted environments. Browser playback depends on reachable endpoints, supported codecs and trusted HTTPS where required.
DEPLOYMENT & CAPACITY
Plan the network.
Size the system.
Start with your camera bitrate, recording retention and viewing patterns. Leave room for the moments when demand rises.
Server starting points
Pilot baselines, not channel capacity guarantees. Validate with your actual cameras before production.
| Role | Pilot baseline | Plan for |
|---|---|---|
| Edge | 4+ CPU cores 4+ GB RAM | Camera LAN, recording disks and selected stream publishing. |
| Cloud | 2+ vCPU 2+ GB RAM | Reachable IP, viewer egress and optional clip storage. |
| Viewer | Separate web node | Reachable Cloud API and media endpoints; playback decoding runs in the browser. |
Ubuntu Server 22.04 / 24.04 · Linux x86-64 or ARM64 · Docker Engine + Compose. Transcoding requires separate CPU/GPU sizing. Network, disk throughput and browser decoding can limit capacity before server memory does.
DEVICE COMPATIBILITY
Keep the cameras you use.
- Hikvision NVRsImport channels and configured names through supported device endpoints. PTZ depends on model, firmware and permissions.
- RTSP / RTSPS camerasConnect reachable camera streams directly. Validate stream URL, authentication and codec.
- LAN discoveryHikvision SADP, ONVIF discovery and RTSP port scanning. Discovery does not guarantee every device feature.
Check the exact NVR model and firmware in a pilot. H.264 is the practical starting point for browser playback; other codecs depend on the browser and playback path.
INTERACTIVE PLANNING TOOL
What does your camera load look like?
Estimate a constant-bitrate workload. Change the values to match the streams you actually publish and record.
Camera ingress / site uplink100 Mbps with headroom
Cloud viewer egress40 Mbps with headroom
Recording storage7.56 TB with headroom
Ingress assumes one published profile per camera and one Cloud destination. Egress counts individual camera plays: four viewers watching four cameras = 16 plays. Additional profiles, Cloud destinations, clip replication, audio and protocol overhead increase usage. Storage uses decimal TB, constant bitrate and the chosen recording hours; headroom is a planning allowance, not a measured overhead. Event recording and variable bitrate change actual results.
How these estimates are calculated
Ingress Mbps = cameras × bitrate. Viewer egress Mbps = concurrent stream plays × bitrate. Storage TB = cameras × bitrate × 1,000,000 ÷ 8 × hours/day × 3,600 × days ÷ 1,000,000,000,000. Planning allowance = estimate × (1 + headroom ÷ 100).
03 / FOR DEVELOPERS
Your application.
Our video layer.
List the available cameras, map stable camera IDs to your interface, and issue scoped playback tokens from your backend. Keep the API key out of the browser.
Read the integration overview01 — LIST AVAILABLE CAMERASBACKEND REQUEST
curl --fail-with-body \
-H "Authorization: Bearer $VIGILSYN_API_KEY" \
"https://console.example.com/api/v1/cameras"CAMERA IDsite-a / entrance
Use the returned ID to connect the right camera to the right part of your application.
Illustrative request. Use your own Cloud address and a scoped server-side key.
04 / OUR STORY
Field experience.
Around 2020.
Good infrastructure starts with understanding the work on the other side of the screen.
Our experience in camera systems began around 2020. Vigilsyn brings that background to a platform built around the practical realities of live video: local networks, distributed sites, recording workflows and applications that need a dependable view.
We believe the infrastructure should be yours to operate. The product should help your team see clearly.
LET’S PLAN THE NEXT VIEW
Tell us about
your deployment.
One site or several. Existing cameras or a new installation. Let’s start with what you need to see.
[email protected]A useful starting point: number of sites and cameras, NVR models, recording needs, and where viewers will connect.