Support

Uniview Human Body Detection Setup Guide

Uniview human body detection can be very useful when a site wants fewer nuisance events than broad motion alone, but the result still depends heavily on the scene, the camera model, the event schedule and the way recording and notifications are linked afterwards.

Uniview Support

Summary

Use this guide when a Uniview site wants better person-focused event behaviour or when the current human-body setup is missing events or producing too many false triggers.

Applies to

  • Supported Uniview cameras and recorders
  • Sites using smart event or human-body style detection
  • EZView and recorder-based notification workflows

Difficulty and time

Difficulty: Moderate

Estimated time: 20 to 45 minutes per camera

What you will need

  • Local access to the camera or recorder
  • Camera model details
  • One clear test scenario using a real person
  • Day and night access if possible

What this guide covers

  • How to confirm feature support and event mode
  • Why scene design matters more than people expect
  • How to tune the area, schedule and linkage
  • How to reduce false alarms without killing the feature

The mistake we see often is enabling a smart feature and expecting it to work perfectly on a poor camera angle. If the camera sees too much road, trees, sky or glare, the menu alone will not rescue the result.

In simple terms, set up the scene first, then the event rule, then the recording linkage, then the phone alert. If you reverse that order, you end up troubleshooting notifications instead of the real event path.

Before you start

Pick one camera and one event objective first. It is much easier to prove one path properly than to make four cameras noisy at once.

  • Confirm the exact camera model supports the feature you want.
  • Check whether the event is configured on the camera, the recorder, or both.
  • Use one real walk-through test rather than guessing from old clips.
  • Note whether the site wants broad awareness or more selective people-only behaviour.
Important

Human body detection is not a cure for bad camera placement

If the camera is aimed across headlights, blowing trees, reflective water or a busy public footpath, false events and missed events become much harder to control.

The cleaner the scene, the better the analytics usually behave.

What usually causes this

  • The camera model does not support the expected smart event.
  • The wrong event type is active and the site is still effectively running broad motion.
  • The schedule or linkage is wrong, so the camera detects but does not record or alert properly.
  • The scene contains too much clutter, glare or irrelevant movement.
  • Night-time behaviour was never tested after day tuning.

Step 1: Confirm the camera actually supports the intended event mode

Before tuning anything, check the exact model and current interface. Some Uniview models support more selective analytics than others, and menu labels can vary by firmware.

  • Open the smart event or analytics section on the camera or recorder.
  • Confirm you are looking at the intended human-body style event and not only general motion.
  • Record the current mode before changing it.
  • If the model cannot do what the customer expects, be honest early rather than over-tuning the wrong feature.

Step 2: Fix the scene before the sensitivity

Installers often go straight to sliders, but the camera view matters more. A badly framed scene creates work for every analytics engine.

  • Reduce unnecessary sky and road if possible.
  • Keep the target path where real people move reasonably central and clear.
  • Watch for tree movement, spider webs, reflective walls and headlights at night.
  • If the camera is too high or too far back, rethink the angle before blaming the analytics.

Step 3: Tune the area, timing and expected target path

Once the scene is reasonable, set the event where people should actually be detected and during the times the site truly cares about.

  • Draw the detection area around the meaningful path, not the whole frame by default.
  • Set the schedule so the event is active during the hours that matter.
  • Walk a real person through the zone to test it.
  • Repeat the test at night if the site needs after-hours protection.

Step 4: Link the event to recording and notifications properly

A good detection rule is still not enough on its own. The system also has to save the right clips and push the right alerts.

  • Confirm the smart event is linked to recording, not just on-screen indication.
  • Check the alert path in the app after the local event has been proven.
  • Use local playback to confirm the clip really exists around the event time.
  • Do not finalise handover until both event detection and evidence capture have been tested.
Worked example

Office side path with false alerts

Situation: A side-path camera kept alerting on tree movement and headlight spill, even though the site only cared about people entering after hours.

Solution used: The camera angle was cleaned up, the target area was narrowed to the walkway, and the rule was tested at night before app alerts were retested.

Why this was chosen: The scene needed work first. Sensitivity changes alone were never going to fix that location.

Installation notes: Local playback after each walk test proved whether the recording linkage was also correct.

Common mistakes

  • Leaving the camera on broad motion and assuming the smart feature is active.
  • Testing only in daylight and ignoring night conditions.
  • Drawing the event area too wide and inviting false triggers.
  • Forgetting to link the smart event to recording and push notifications.

Troubleshooting table

Symptom What to check What to do next
Too many alerts Scene clutter, detection area, wrong event type Reduce the scene and target area before pushing sensitivity changes too far.
No useful clips from detected events Recording linkage, schedule, playback search Confirm the event is actually linked to recording and review playback locally.
Works by day but not at night Lighting, reflections, angle, night settings Retest the rule after dark and adjust the scene or zone accordingly.
Customer expects more than the model can do Camera capability and analytics support Check the datasheet and reset expectations before more tuning.

When to contact support

Contact SecurityWholesalers support with your order number, camera model, firmware and a clear description of the event problem if you still need help. It is especially useful to include screenshots of the event area and one example of a false or missed trigger.

Related support guides

Related buying guides

Relevant product categories

Still stuck?

Need help choosing or setting up a system? Contact SecurityWholesalers support with your order number, product model and a clear description of the issue.

Frequently asked questions

  • What is Uniview human body detection meant to do?

    It is meant to be more selective than broad motion on supported models, especially where the site wants people-focused events.

  • Why am I still getting false alarms?

    Usually because the scene, angle or detection area still needs work.

  • Do I need to set recording linkage separately?

    Usually yes. The event has to be linked to recording and alert paths properly.

  • Should I test day and night separately?

    Yes. Night conditions often change analytics behaviour more than people expect.

  • What should I send support?

    Send the camera model, firmware, screenshots of the event area, and one example of a false or missed trigger.

Further Reading
Overview Overview Guide Page Uniview Support Guides Guide Page Hikvision Support Guides Guide Page HiLook Support Guides

Wrap up EOFY with extra bonus savings :) Use Coupon code: EOFY26 on orders over 800.
Trade Customers: Log In or Register to Unlock Even Better Prices.

Save & Share Cart
Your Shopping Cart will be saved and you'll be given a link. You, or anyone with the link, can use it to retrieve your Cart at any time.
Save Cart & Generate Link
Send Cart in an Email Done! close
Empty cart. Please add products before saving :)

Back Save & Share Cart
Your Shopping Cart will be saved with Product pictures and information, and Cart Totals. Then send it to yourself, or a friend, with a link to retrieve it at any time.
Send Cart Email
Your cart email sent successfully :)