One platform to search, review, and create court-ready reports for video.

Designed for law enforcement, AI-powered analytics turn video into actionable intelligence faster.

17 Years
Serving police forces and national agencies worldwide
CCTV/DVR formats handled, solved at ingest
AI Trust
Investigation focused layered analytics you can trust
ISO 17025
Aligned to workflow, automatic audit trail and disclosure logs
Read the transcript

Video is one of the most powerful tools in modern policing. The challenge is finding what matters. Kinesense uses intelligent video analysis to cut through the noise, so you no longer have to watch endless hours of footage. Set up real-time alerts that trigger when a license plate or a person of interest is recognized in a live video stream, or for post-event investigations, our system can ingest almost any video format and automatically find the key events. Search for vehicles by type, color, or even a partial license plate. Find people using detailed attributes such as clothing, headgear, or backpacks, all powered by carefully curated datasets, with performance, reliability, and compliance at its core. In policing, time matters, but so does trust.

Kinesense delivers a secure end-to-end workflow, empowering investigators to unlock the full potential of their video evidence. We provide powerful transparent analysis with complete audit trails. No black boxes, just courtroom-ready insights fast.

Any footage in. Defensible evidence out.

Kinesense analyses frames on CPU or GPU, and returns the search results, alerts, reports and an unbroken chain of evidence.

CAN INGEST VIDEO FORMATS

Kinesense ingests 3000+ video formats, including MP4, AVI, MOV, MKV and MTS.

LIVE STREAMS

Kinesense also ingests live streams from CCTV, body-worn, mobile and network cameras.

•••
  • EVENT DETECTION
  • ATTRIBUTE RECOGNITION
  • ANPR CAPABILITIES
  • REDACTION
GPU or CPU

Works on both

SUPPORTS 2FPS & ABOVE

TRAINED ON REAL POLICE FOOTAGE

SEARCH RESULTS

Kinesense surfaces the search results in a recording, each stamped with its timecode.

ALERTS

Red SUV: Vehicle match

Camera 12

01:07:18

HIGH

REPORTING

Total Videos Processed

247

Events Detected

1,286

CHAIN OF EVIDENCE

Integrity Verified & Protected

  1. Original
    File
  2. Ingested
    (Hash)
  3. Analysed
    (Results)
  4. Exported
    (Report)
  5. Archived
    (Secured)

AI you can explain. Intelligence you can defend.

Kinesense combines decades of explainable machine vision with AI trained on real investigative footage to deliver defensible intelligence.

01FOUNDATION LAYER

Human-engineered machine vision

Classifier-based algorithms, rule-based and deterministic. Their behaviour can be written into a statement and defended under cross-examination; no training data to account for, no drift.

  • EVENT DETECTION
  • EVENT CLASSIFICATION
  • DIRECTION, COLOR, & ZONE BASED ANALYTICS
02AUGMENTATION LAYER

Deep learning trained on police video

Trained on anonymised footage representative of real investigations: night-time and infrared imagery, reduced image quality, heavy compression, awkward angles, and adverse weather conditions. Not scraped from the internet.

  • PEOPLE & VEHICLES
  • OBJECT ATTRIBUTES
  • LICENCE PLATE DETECTION & RECOGNITION
WHAT THE TWO PRODUCE TOGETHER

An investigator can search using rich descriptors but work back to robust event detection, giving confidence that they are not missing something that they later have to testify to on court.

Explainable, Not Opaque

Analytics are designed so investigators can understand why an event was identified, not simply accept an unexplained confidence score.

Trained on Police Reality

Deep learning models trained on footage typical of real investigations: night-time, degraded image quality, heavily compressed, and off-angle.

Reliable in Bad Conditions

A layered approach combines tried-and-tested motion and event detection with deep learning for more detailed analysis.

CPU or GPU: You Decide

Run standard tasks on existing CPU hardware. Leverage available GPU hardware for accelerating advanced analytics. Works out-of-the-box on almost any GPU.

What Our Customers Say

Feedback from professionals who rely on Kinesense for critical investigations.

A six-week viewing job came back in two days, with the exhibit list already written.
Detective SergeantMajor Investigation TeamGreater Manchester Police
It was down to the purchase and use of the system that the investigation team was able to accurately provide evidence of offending in the biggest investigation Kent Police has ever embarked on. Considerable cost-saving was achieved.
Assistant Chief ConstableKent Police
I was really impressed with the speed of service and the product delivered. It organised my 2.5 million files down to 950 files that the case team were able to view easily and quickly found footage of evidential value.
Senior InvestigatorInvestigation LeadHMRC

Frequently Asked Questions

Everything you need to know about deploying and using the platform.

Is Kinesense a BriefCam alternative?

Yes, for investigations. Both accelerate video review, but they are built for different end points. Analytics platforms are optimised for large camera estates and operational dashboards; Kinesense is optimised for the exhibit: ingest of proprietary DVR formats, documented clarification, redaction, disclosure logs and an audit trail that goes to court. If your requirement ends in a report served on the defence rather than a dashboard, evaluate us.

Are your analytics a black box?

No. Tried-and-tested classical computer vision algorithms underpin our analysis, with deep learning layered on top for attribute recognition. This provides investigators with information about what was detected, which processing layer identified it, and where it occurred in the footage, not simply an unexplained confidence score.

Do we need GPUs to run Kinesense?

No. Motion detection and object classification run on CPU, on hardware most units already own. GPU is recommended and only needed for the in-depth attribute layers. That choice matters when the budget is a business case, and when the position you are deployed to has limited power. If you have a GPU, use it, and ask us for sizing figures for both.

Does using your tools affect admissibility?

Not when the original is preserved and every processing step is recorded. Our solution comes with a complete disclosure log: what video was imported, by whom and when; how it was watched, with what algorithm, and whether on a frame-by-frame basis; who clarified it and how; and what reports were generated and when.

How is Kinesense licensed?

Modularly. Start with the capability the caseload demands and add modules later, rather than buying a suite to reach one feature. We price on an annual subscription model, which means you get access to the latest version.

See all FAQs

Accelerate Truth. Close Cases Faster.

Join 40+ police forces and security organisations using Kinesense to solve cases faster and more efficiently.

Format support

3000+ video formats

Anything playable in Windows Media Player, VLC or FFmpeg is supported, without the manufacturer's original player. Have a format that isn't listed? Our engineers have 17 years experience of dealing with difficult formats.