EDGE AI & ATTENTION MANAGEMENT

Hundreds of Cameras.
One Human Can’t Watch Them All.

Azrian Guard uses AI to detect activity, prioritize the camera that needs attention, and alert the right person, working seamlessly with your existing surveillance infrastructure.

Customer-controlled video
Setup in minutes
Works with existing cameras
Azrian Guard (Master Node)
Online (Master Node)

More Cameras Don’t Create
More Human Attention.

A security operator cannot continuously monitor dozens or hundreds of feeds. Most surveillance video is completely uneventful. The real challenge in physical security isn’t recording footage, but knowing where to look the moment something happens.

AI doesn’t replace the guard. It directs the guard’s attention, elevating critical activity straight to the foreground.
01 • Detect

AI Identifies Activity

Motion pre-filters and neural AI analyze scene dynamics, ignoring static empty footage while isolating people and vehicles.

02 • Prioritize

Camera Enters Attention

When an event occurs, the affected camera dynamically jumps to the top-left priority anchor tile on your video wall.

03 • Alert

Operator Alerted Remotely

The system triggers local control-room chimes and dispatches instant encrypted push notifications to on-duty mobile guards.

04 • Review

Pre/Post Incident Context

Operator reviews seamless video recorded with 3s pre-event buffer to immediately understand how the incident began.

05 • Decide

Human Security Decision

The human makes the final call, verifying whether the alert is an actual threat, dispatching patrol, or sounding sirens.

Real-World Case Concrete Operational Scenario:

A 64-camera facility generates thousands of hours of video, most of which contains no security-relevant activity. When an unauthorized person approaches Gate 4 after hours, the operator doesn’t need to scan across an 8×8 grid of tiny video tiles. Azrian Guard immediately elevates Gate 4 to the primary display tile, sounds a control-room alert, records the incident with pre-event context, and pushes the event clip directly to mobile security.

Works With What You Already Have

Upgrade your intelligence layer without throwing away functional equipment or paying for proprietary cloud cameras.

Existing IP Cameras & NVRs

Connects directly to any standard RTSP or ONVIF stream. Built-in profiles for Hikvision, Dahua, Uniview, Reolink, Tapo, and multi-channel NVR recorders.

Standard PCs & NVIDIA GPUs

Zero proprietary server appliances required. Native C++ engine with MotionGate smart pre-filtering runs efficiently on standard Windows workstations and consumer or commercial GPUs.

Private, Local-First Architecture

Video streams, AI inference, recordings, and incident logs remain on customer-controlled infrastructure. No centralized surveillance-video cloud required.

Why Edge AI Outperforms Cloud & Legacy NVRs

Private local edge processing vs. bandwidth-intensive cloud relay and unmanaged standalone recorders.

Capability Cloud AI Surveillance Traditional NVR
Azrian Guard Azrian Guard
Edge AI
Camera Ingestion
Works with existing cameras
Standard RTSP only
Proprietary cameras preferred
Standard RTSP only
No native AI ingestion layer
Universal RTSP & ONVIF
Zero hardware replacement required
Video Storage Location
Where recordings reside
Centralized cloud servers
Recurring monthly storage invoices
Customer local storage
Unmanaged single hard drive
Customer-Controlled Local Storage
Zero-CPU stream copy remuxing
AI Inference Execution
How models process frames
Cloud data centers
Requires high outbound upload
Basic pixel motion
High false alarms from headlights & rain
Local Edge AI (PC or GPU)
Real-time YOLOX neural inference
Cluster Architecture
Handling dozens of cameras
Cloud managed
Choked by internet uplink limits
Standalone recorder
Hard ceiling on per-box compute
Master / Worker Cluster
Parallel load-balancing across auxiliary PCs
Guard Attention
How events are presented
Dashboard notification feed
Does not reorder active video tiles
Static unprioritized grid
Guards must scan across 64+ feeds
Dynamic Anchor Tile Elevation
Active threat jumps to top-left anchor tile
Remote Networking
Remote access security
Cloud video relay
Surveillance routes through cloud servers
Inbound port forwarding
Exposes firewall to public network scans
Encrypted Peer-to-Peer Mesh
Zero port forwarding via WireGuard mesh
Supported Brands: Hikvision • Dahua • Uniview • Reolink • TP-Link Tapo • Axis Communications • Bosch • Xiongmai • Universal RTSP • ONVIF Profile S/T

Distributed AI Processing Across Multiple Machines

Master and Worker nodes coordinate as a unified private cluster. Scale inference across auxiliary PCs and GPUs without complex server licenses.

1. Existing Cameras RTSP/ONVIF streams from current IP cameras or NVRs
2. Master Node Local coordinate server, recording buffer & storage manager
3. AI Worker Nodes Auxiliary PCs & GPUs running parallel YOLOX inference
4. Attention Priority Top-left Anchor Tile elevation + mobile encrypted push
5. Human Decision Guard confirms threat and coordinates immediate action

Built for Real-World Deployment

Runs on standard Windows hardware with flexible GPU acceleration, customer-controlled video, and local storage.

AI Detection & 4-Point Zones

Fast person and vehicle detection with local YOLOX neural models. Draw custom 4-point quadrilateral boundaries on live video to isolate gates and driveways while ignoring public sidewalks.

Auto-Discovery & NVR Importer

Zero-config ONVIF WS-Discovery and subnet scanner finds cameras automatically. Batch-generate up to 64 NVR channels from a single IP with mainstream and substream RTSP routes.

Master & Worker Nodes

Distribute AI processing across available hardware as your camera network grows. Master and Worker nodes form a distributed system, streaming frames and load-balancing inference over private networks.

Direct Stream Copy Recording

Direct remuxing saves high-definition H.264/H.265 continuous and event video segments with no CPU/GPU video re-encoding, backed by fail-safe clip finalization.

Mobile Alerts & Local Sirens

Trigger local hardware sirens instantly upon verified entry, and receive encrypted push notifications and video clips on Android and iOS devices.

Multi-Monitor Video Wall

Detach and distribute camera grids across secondary monitors. Dense 1x1 to 8x8 layouts with auto-prioritization dynamically bring active feeds front and center.

AI Schedules & Dwell Time

Schedule AI inference per camera by day of week and operating hours. Suppress transient pedestrians with configurable 5s–60s dwell time loitering filters.

Multi-Brand Compatibility

Pre-configured profiles for Hikvision, Dahua, Uniview, Reolink, Tapo, Xiongmai, port-mapped NVRs, and mobile webcams (DroidCam, IP Webcam).

Private, Local-First Architecture

Video streams, AI models, and incident databases remain on customer-controlled infrastructure. No centralized surveillance-video cloud required.

Simple, Transparent Pricing

A single one-time fee for the desktop software, plus a scalable subscription for your cameras.

VOLUME DISCOUNT PRICING TIERS
Base Software License: $50 USD • One-Time
DEPLOYMENT TIER CAMERAS MONTHLY ANNUAL (PER CAM) SAVING / STATUS
Tier 1 (Starter) 1–4 $11.99/cam $119/cam Save 17%
Tier 2 (Growth) 5–9 $10.49/cam $99/cam Save 21%
Tier 3 (Commercial) 10–24 $8.99/cam $89/cam Save 17%
Tier 4 (Enterprise) 25–49 $7.99/cam $79/cam Save 18%
● Tier 5 (Fleet / Scale) 50+ $6.49/cam $59/cam Current Plan
HIGH-CAPACITY FACILITY PACKAGES Distribution centers · Multi-building networks

100-Camera Package

Up to 100 cameras • $590 / mo

or $5,900 / yr billed annually

Request 100-Cam Pilot

200-Camera Package

Up to 200 cameras • $1,180 / mo

or $11,800 / yr billed annually

Request 200-Cam Pilot

200+ / Multi-Site

Custom Volume Pricing

Dedicated cluster engineering

Contact Enterprise Sales
Open Interactive Quota Calculator & Custom Slider →
Transparent pricing No hidden cloud fees Cancel anytime Secured by Stripe
Early Access & Pilot Program

Have an Existing Camera Network?

We are partnering with commercial facilities, warehouses, and security operations in Melbourne and Australia to evaluate Azrian Guard on real surveillance infrastructure with dedicated engineering support.

Request a Pilot See How It Works
SYSTEM CAPABILITIES

Everything You Need for
Smarter Video Analytics

Designed for high-throughput edge execution and built for real-world deployment. Explore the architectural capabilities powering Azrian Guard's distributed video analytics pipeline.

PILLAR 1 • OPERATOR EXPERIENCE

What the Security Operator Gets

01 • Targeted YOLOX AI Detection

Accurate person and vehicle detection executing locally on the edge. Runs continuously across feeds to isolate human and vehicular activity without latency.

02 • Dynamic Top-Left Prioritization

When an event occurs, the affected camera dynamically jumps to the top-left priority anchor tile on your video wall, ensuring guard attention is never split across a dense grid.

03 • 4-Point Detection Zones (ROI)

Right-click any camera to draw a custom 4-point quadrilateral zone. Restrict alerts to sensitive entryways while ignoring public sidewalks, streets, and swaying trees.

04 • Pre & Post Incident Buffering

Captures 3 seconds of circular pre-alarm footage plus 3 seconds of post-alarm footage. Every incident clip shows how the intruder arrived before the alarm sounded.

05 • Mobile Alerts & Dial-Out Response

Review live streams and bounding boxes on iOS/Android. In the event of a verified intrusion, operators can trigger emergency contacts with a single tap.

06 • Multi-Monitor Video Wall

Spawn independent satellite display windows across secondary monitors or commercial video walls, supporting 1x1, 2x2, 3x3, 4x4, 6x6, and 8x8 layouts.

PILLAR 2 • IT & INFRASTRUCTURE

What the IT Team Gets

01 • Works With Existing Cameras & NVRs

Universal RTSP and ONVIF Profile S/T compatibility. Includes automated WS-Discovery plus a one-click NVR batch importer for up to 64 channels.

02 • Customer-Controlled Video

No centralized video storage cloud. Raw streams, neural inferences, database audit logs, and recordings stay strictly on customer-controlled infrastructure.

03 • Master / Worker Cluster

AiWorkerPool streams raw frames to worker machines in parallel, allowing facilities to distribute compute load across multiple PCs as camera counts scale.

04 • Encrypted Peer-to-Peer Networking

Integrates with Tailscale mesh VPN for secure multi-building or remote worker connectivity without opening inbound firewall ports.

05 • Direct Stream Remuxing

Direct stream remuxing (-c:v copy) writes H.264/H.265 video packets into MP4 files with no CPU/GPU video re-encoding.

06 • Embedded SQLite & Rolling Storage

Embedded SQLite database manages configurations, schedules, and incident metadata locally, while automated storage quotas safely prune oldest files.

PILLAR 3 • DEEP ENGINEERING

Technical Architecture & Optimization

MotionGate EMA Pre-Filter

An Exponential Moving Average filter pre-checks each frame. When scenes are static, AI inference is skipped entirely, saving 20 to 30 ms per frame.

Dual-Stream Decoupled Processing

Substream (low-res) runs AI inference and grid previews to preserve framerate, while Mainstream (high-res) is captured for sharp incident recordings.

Emergency Fail-Safe MP4 Finalizer

If network drops or power is cut during an active incident, an automated recovery routine cleanly finalizes MP4 headers to help preserve recoverable incident clips.

Time-Based Operating Schedules

Set per-camera active hours and days of the week. Restricts heavy inference to after-hours surveillance windows and conserves compute during the day.

Hardware Utilization & VRAM Quotas

Supports CPU-only and GPU mode (CUDA/TensorRT). Explicitly set GPU VRAM quotas and thread worker pools so the system never starves the host OS.

Transport Tuning & Dynamic FPS

Select TCP or UDP per camera to eliminate packet drop and video smearing, with real-time FPS throttling (1–30 FPS) based on motion state.

Performance Engineering

Tested for 40+ Concurrent Dual Streams

Thanks to lightweight MotionGate pre-filtering, Azrian Guard ignores static empty frames and limits heavy AI processing to true motion events. Tested for 40+ concurrent streams on selected hardware configurations (e.g., modern multi-core x86 with dedicated GPU acceleration). Actual capacity depends on camera resolution, codec, FPS, GPU, and AI configuration.

TRANSPARENT PRICING

Simple, Transparent Pricing
That Scales With You

Predictable per-camera licensing with zero hidden cloud fees, zero vendor lock-in, and all AI features included.

VOLUME DISCOUNT PRICING TIERS
Base Software License: $50 USD • One-Time
DEPLOYMENT TIER CAMERAS MONTHLY ANNUAL (PER CAM) SAVING / STATUS
Tier 1 (Starter) 1–4 $11.99/cam $119/cam Save 17%
Tier 2 (Growth) 5–9 $10.49/cam $99/cam Save 21%
Tier 3 (Commercial) 10–24 $8.99/cam $89/cam Save 17%
Tier 4 (Enterprise) 25–49 $7.99/cam $79/cam Save 18%
● Tier 5 (Fleet / Scale) 50+ $6.49/cam $59/cam Current Plan
HIGH-CAPACITY FACILITY PACKAGES Distribution centers · Multi-building networks

100-Camera Package

Up to 100 cameras • $590 / mo

or $5,900 / yr billed annually

Request 100-Cam Pilot

200-Camera Package

Up to 200 cameras • $1,180 / mo

or $11,800 / yr billed annually

Request 200-Cam Pilot

200+ / Multi-Site

Custom Volume Pricing

Dedicated cluster engineering

Contact Enterprise Sales
CUSTOM STREAM ESTIMATOR

Interactive Camera Quota Calculator

Drag to calculate your exact investment based on active stream count.

Cameras
1 Cam 25 Cams 50 Cams 75 Cams 100+ Cams
Selected Tier Tier 2 (Growth) 5–9 cams
Rate per Camera $10.49 USD/mo Volume rate
Software License $50 USD One-time node fee
Recurring Total $83.92 USD/mo Subscription
Total Due Today $133.92 USD Base fee + streams
Checkout with Stripe
Transparent pricing No hidden cloud fees Cancel anytime Secured by Stripe

Frequently Asked Questions

Can I use my existing cameras and NVRs?

Yes! Azrian Guard works with any standard IP camera or NVR that outputs an RTSP/HTTP stream (including Hikvision, Dahua, Uniview, Reolink, Tapo, Xiongmai, port-mapped recorders, and mobile webcams). The built-in network scanner auto-discovers ONVIF cameras on your LAN, and the NVR wizard allows batch-importing up to 64 channels from a single IP in one click.

Can I limit alerts to specific perimeters or zones?

Yes. With 4-Point Detection Zones (ROI), you can right-click any camera feed to draw a custom quadrilateral boundary around sensitive gates, doors, or driveways. The AI tracks all targets on screen, but only triggers audible sirens, mobile alerts, and event recordings when targets penetrate inside your defined zone, filtering out busy sidewalks and public streets.

How does Azrian Guard record and manage video?

Video is saved directly to local storage on your Master node using zero-CPU stream copying (-c:v copy) and an embedded SQLite database, with zero cloud bandwidth fees and zero transcode overhead. Every alert captures a seamless clip with a 3-second pre-recording and 3-second post-recording buffer, backed by automatic fail-safe header finalization.

How can I scale to dozens or hundreds of cameras?

For large facilities, Azrian Guard supports a Master/Worker cluster architecture and multi-monitor video walls. Worker nodes can be added across your local network or remotely via Tailscale. The Master node delegates AI processing across the worker pool and lets you distribute 4 to 64+ camera tiles across separate physical control room displays.

Is there a mobile app and remote access?

Yes! We offer a companion iOS and Android app. Using Tailscale's encrypted peer-to-peer WireGuard mesh, your app connects securely to your Master node without opening firewall ports. In the event of a threat, the app allows you to review live streams and instantly dial emergency numbers with a single tap.

Is this difficult for our IT team to manage?

Not at all. Get a basic installation running in minutes. Enterprise and multi-site deployments may require additional network and hardware configuration. We also embedded a dedicated IT console directly in the UI, allowing system administrators to inspect logs, manage Tailscale connections, configure GPU VRAM limits, set time-of-day AI schedules, and debug instantly.

DOCUMENTATION

Quick Start Guide
Setup & Distributed Cluster

How to install Azrian Guard, configure Tailscale peer-to-peer mesh, and connect distributed Worker Nodes.

Part I • Quick Start Guide

Follow steps 1–4 to install, connect cameras, configure your perimeter zone, and start monitoring. Get a basic installation running in minutes; multi-site deployments may require additional network planning.

1. Installation & Activation

  1. Download the AzrianGuard_Client.zip file from the Downloads page.
  2. Right-click the zip file and extract it to a permanent location (e.g., C:\AzrianGuard).
  3. Run AzrianGuard.exe as an Administrator.
  4. Upon first launch, you will be prompted to Sign Up. Enter your Company Name, Full Name, Country, and create a Password to establish your administrative account.
  5. After completing the sign-up, you will be asked for your license key. Paste your ASRIAN-XXXX key to officially activate your Master Server.

2. Adding Cameras & Network Discovery

Azrian Guard connects natively to IP cameras, NVRs, and RTSP/HTTP streams without vendor lock-in. You have three flexible methods to add streams:

Method A: Automated Network Discovery (ONVIF & Subnet Scanner)

Click "Discover Cameras" in the top navigation bar. Azrian Guard broadcasts ONVIF WS-Discovery probes (UDP multicast 239.255.255.250:3702) and rapidly scans common surveillance ports (554, 80, 8000, 8080, 8554, 5543). Click "Test Stream" on any discovered device to measure latency and verify credentials before adding it to your system.

Method B: NVR / DVR Multi-Channel Batch Importer

If your cameras are connected through an NVR or DVR, click "Import NVR Channels". Enter the recorder's single IP address, credentials, starting channel, and channel count (4 to 64). Select your brand format (Hikvision ch101/ch102, Dahua subtype=0/1, Uniview s0/s1, Tapo, Xiongmai, or Port-Mapped sequential ports). Azrian Guard auto-generates both Mainstream and Substream URLs for all channels in one click.

Method C: Manual RTSP / HTTP Stream

Click "Add Camera". Enter your custom RTSP stream URL (e.g., rtsp://<username>:<password>@<camera-ip>:554/<stream-path> or local camera endpoints such as http://<camera-ip>:8080/stream.sdp). Specify optional Substream URLs to maximize multi-camera grid rendering framerates.

3. Configuring 4-Point Detection Zones (ROI & Perimeter)

Eliminate nuisance alerts from public roads, sidewalk pedestrians, and swaying foliage by defining custom 4-point quadrilateral detection perimeters on any camera:

  1. Right-click any camera tile on the live dashboard grid and select "Configure Detection Zone".
  2. A high-resolution snapshot dialog appears with 4 draggable corner handles (numbered 1, 2, 3, 4).
  3. Click and drag the corner handles to outline your secure perimeter (e.g., driveway boundary, front porch, entry gate, or inventory aisle).
  4. Ensure the "Enable Detection Zone Filtering" checkbox is checked and click "Save Zone".
  5. The camera tile will display a glowing perimeter boundary and a ZONE ACTIVE badge. The AI will continue tracking all targets on screen, but alarms and sirens will only trigger if a target penetrates inside your 4-point polygon!

4. Multi-Monitor Video Wall & Satellite Displays

Azrian Guard is built for command centers and security operations rooms requiring multi-display monitoring:

  • Click "Video Wall" from the top toolbar to spawn an independent Satellite Display window.
  • Drag the window onto secondary or tertiary physical monitors or commercial video wall arrays.
  • Select custom grid layouts (1x1, 2x2, 3x3, 4x4, 6x6, or 8x8) and assign specific camera groups or pagination per display.
  • Dynamic Auto-Prioritization: When an intrusion or tripwire alert fires on any camera, that camera immediately jumps to the top-left priority tile on your main display for instantaneous visual verification.
Part II • Advanced Cluster & Network Infrastructure

Configure private mesh connectivity, distributed worker load balancing, and GPU acceleration for high-camera deployments.

5. Remote Access & Multi-Site Networking (Tailscale)

Because Azrian Guard uses a Private, Local-First Architecture, surveillance video and AI processing remain strictly on your local machines and are not uploaded to centralized cloud video storage. If you want to securely view your cameras from home, or connect Worker Nodes from other buildings, we recommend using Tailscale (an encrypted mesh VPN with simple setup).

Step 1: Download Tailscale Click here to download Tailscale (Free)
  1. Install Tailscale on the computer running your Master Server. Log in to create your secure network.
  2. Install Tailscale on your personal laptop or mobile device and log in with the same account.
  3. Open the Tailscale app to find your Master Server's secure IP address (it will start with 100.x.x.x).
  4. You can now securely access the Azrian Guard dashboard remotely using that IP, without exposing your cameras through public port forwarding.

6. Connecting Worker Nodes (Load Balancing Cluster)

If you have purchased a large camera license (e.g., 30+ cameras) and want to split the AI processing load across multiple computers:

  1. Download the exact same AzrianGuard_Client.zip on your second computer (the Worker Node).
  2. Ensure both computers are on the same local network, or connected via Tailscale.
  3. Launch AzrianGuard.exe on the Worker Node. Do not enter a license key.
  4. Click "Connect to Existing Master Server".
  5. Enter the local IP address (or Tailscale IP) of your Master Server.
  6. The Master Server will automatically authorize the Worker Node and begin delegating AI processing tasks to it. No secondary license key is required!
Important: Remote Workers & Private Camera Networks

If your Worker Node is in a completely different building (connected via Tailscale), but your cameras are on the Master Server's private local network (e.g., 192.168.1.x), the Worker will fail to reach the camera's RTSP stream. To fix this, you must configure Tailscale on the Master Server as a Subnet Router:

  1. On the Master Server, open Command Prompt as Administrator.
  2. Run the command: tailscale up --advertise-routes=192.168.1.0/24 (replace with your camera's actual subnet).
  3. Log into your Tailscale Admin Dashboard online, click on your Master Server's machine settings, and Approve the subnet route. Your remote Worker Node will now be able to transparently process your local cameras!

7. Mobile App Access & Approval

You can view live feeds and receive security alerts directly on your smartphone using the Azrian Guard mobile app. For maximum security, mobile access is restricted by a strict peer-to-peer approval system:

  1. Download the Azrian Guard mobile app and the Tailscale mobile app onto your phone.
  2. Open Tailscale and log in using the same network account as your Master Server.
  3. Open the Azrian Guard mobile app. Enter your Name and your Master Server's Tailscale IP address (100.x.x.x).
  4. This will send a secure connection request directly to the Master Server.
  5. An administrator must open the Azrian Guard desktop app on the Master Server and manually Approve your mobile device's request.
  6. Once approved by the Master, configure your notification settings in the mobile app and you are good to go!

8. Enabling NVIDIA GPU Acceleration

Azrian Guard supports NVIDIA GPUs (CUDA/TensorRT) for significantly faster AI inference and camera processing. To enable this:

  1. Ensure your NVIDIA graphics drivers are fully updated (download the latest Studio or Game Ready drivers directly from NVIDIA).
  2. Open the Azrian Guard desktop application and navigate to the Settings page.
  3. Find the Hardware Acceleration section and attempt to connect/enable the NVIDIA GPU.
  4. Important: Once it successfully connects, you must completely close the app and open it again (restart) for the hardware acceleration to take effect on your camera streams.

9. Advanced Settings & AI Schedules

Azrian Guard is designed to be incredibly powerful, but never hard to use. Once your Master Server is running, open the Settings page in the desktop application to fine-tune your exact setup:

  • Time-Based AI Operating Schedules: Configure active hours and active days of the week per camera so high-compute neural inference automatically arms overnight and rests during office hours.
  • Dwell Time / Loitering Filters: Set a custom 5s–60s minimum presence duration per camera to suppress transient pedestrians while alerting on persistent loiterers.
  • Motion & AI Thresholds: Fine-tune confidence cutoffs and motion gate sensitivity on a per-camera basis.
  • Stream Copy Recording & Rolling Storage: Specify local drive storage quotas and enable direct stream copying with no CPU/GPU video re-encoding.
  • Tactical Themes & Grid Layouts: Toggle between Dark and Light tactical themes and customize default multi-camera grid densities (1x1 to 8x8).

10. Need Help?

Having trouble setting things up?

If your cameras aren't connecting, Tailscale is giving you trouble, or anything else isn't working as expected, please let us know!

Contact Support
OUR MISSION & STORY

Smarter Security Built on
Privacy, Performance & Freedom

We built Azrian Guard around a clear observation: more cameras don't create more human attention. We provide organizations with intelligent attention filtering while keeping video strictly under customer control.

Customer-Controlled Video

Your video feeds belong to you. Azrian Guard processes AI inference and recordings locally on your edge hardware. Surveillance footage is never routed to centralized cloud video storage, eliminating bandwidth choke points and cloud privacy exposure.

Zero Hardware Lock-In

Universal compatibility with any standard RTSP or ONVIF camera stream. Keep your existing camera investments and upgrade your intelligence layer without expensive hardware replacements.

Directed Attention Management

We solve operator fatigue. Instead of forcing guards to watch dozens of static feeds, dynamic auto-prioritization elevates active perimeter intrusions to the top-left priority tile on your video wall for instantaneous visual verification.

High-Throughput Edge Performance

Through native C++ engineering and lightweight MotionGate pre-filtering, Azrian Guard delivers fast AI analytics on standard Windows PCs and selected GPU configurations without requiring specialized server racks.

Distributed Cluster Scalability

Scale across large commercial facilities. Master and Worker nodes coordinate as a unified private cluster, delegating heavy AI inference across auxiliary PCs and GPUs over private mesh networks without per-node server fees.

Permissive Commercial Licensing

Our neural inference pipeline uses YOLOX under the permissive Apache-2.0 license, supporting commercial deployment and integration.

Surveillance Is an Attention Problem

We started with a simple observation: more cameras don’t create more human attention. Modern commercial facilities can easily deploy dozens or hundreds of cameras, generating thousands of hours of video where most of the footage contains no security-relevant activity.

The real challenge for a security team isn’t storing another terabyte of video or replacing functional IP cameras. The challenge is helping a human operator understand what deserves attention right now.

Azrian Guard was built specifically around that problem. By coupling targeted edge AI filtering with dynamic video wall prioritization, we elevate critical events to the human operator the second they happen, while keeping video streams, recordings, and infrastructure entirely under customer control.

Speak with Our Engineering Team

For commercial deployments, multi-site architecture questions, or custom integrations:

✉ Contact Engineering & Sales
OFFICIAL RELEASE

Download Azrian Guard
Version 2.0.0 for Windows

Self-contained Windows 64-bit client bundle with high-throughput Qt 6 runtime, ONNX Runtime GPU inference libraries, and pre-packaged YOLOX neural weights.

Production Release x86_64 Architecture

Azrian Guard Desktop (v2.0.0)

OS: Windows 10 / 11 (64-bit) • File Size: 111.2 MB • License: Commercial

Includes full desktop GUI binary (AzrianGuard.exe), native FFmpeg stream decoders, ONVIF discovery scanner, NVR channel generator, 4-point perimeter zones, CUDA 11/12 GPU provider libraries, and MotionGate pre-filter engine. No cloud-based video processing or cloud-hosted NVR required.

Downloads are locked

You must have an active Azrian Guard license key to download the software.

Purchase a License
Or

Already have a license key?

Azrian Guard Logo PORTABLE WINDOWS APP No traditional installer required

Rapid Installation in 3 Steps

Get up and running in 15 minutes with zero proprietary hardware configuration.

01

Extract Application Bundle

Extract AzrianGuard_Client.zip into your preferred directory (e.g., C:\AzrianGuard). No invasive installer or registry bloat.

02

Prerequisites & Drivers

Ensure the Microsoft Visual C++ 2015–2022 Redistributable is installed. For GPU acceleration, verify your NVIDIA driver is updated.

03

Launch & Ingest Cameras

Launch AzrianGuard.exe. Enter your camera RTSP addresses or scan via ONVIF to start real-time threat detection.

GET IN TOUCH

Connect with Our
Engineering & Solutions Team

Have questions about scaling across multiple facilities, hardware sizing, or custom neural models? We're here to help you get the most out of Azrian Guard.

Direct Support

For technical inquiries, license quota extensions, and implementation guidance:

contact@azrianguard.com

System Sizing & Benchmarks

Need help sizing GPU and CPU hardware for 50+ camera streams? Contact our engineering team for an exact capacity plan.

Commercial & Industrial Facilities

Designed for commercial properties, storage facilities, industrial warehouses, and multi-building campuses with customer-controlled infrastructure.

Send a Message

Fill out the form below and we usually respond within one business day.

SECURE CHECKOUT

Complete Your Subscription

Enter your details to finalize your plan. Payments are processed securely via Stripe.

Contact Information

Payment Details

Guaranteed safe & secure checkout powered by Stripe.

Order Summary

Growth Tier
$10.49 USD / camera / mo
Tier 2 (Growth) 8 Camera Licenses (Monthly)
$83.92
Base Software License One-time setup fee
$50.00
Subtotal $133.92 USD
Applicable Tax Calculated by Stripe at checkout
Total due today $133.92 USD + applicable tax
  • • High accuracy person & vehicle detection
  • • Encrypted peer-to-peer networking
  • • Customer-controlled video & local storage