CogVisIntelligence
Real-time GPU vision: plates and faces.
In Development
Teaching machines to lookComputer vision
Detection, tracking, recognition and OCR — running on your hardware.
Real-time vision systems: object detection and tracking, licence plate and face recognition, OCR through degradation, hazard detection, and pose or depth estimation, deployed on the hardware you actually have.
How we approach it
CogVisIntelligence reads plates from any angle through extreme degradation and tracks them with ghost-resistant persistent IDs. FireSmokeDetector runs a three-tier pipeline — a fast detector proposes, a vision-language model confirms the context, depth estimation grounds the scale — because in that domain a false positive is expensive and a miss is catastrophic.
Both run locally through ONNX Runtime with CUDA. That is usually the deciding constraint: the sites that need this cannot send frames to somebody else's cloud, for privacy, for bandwidth, or because the link goes down and the camera does not.
You get a measured false-positive and false-negative rate on your own footage, not a vendor benchmark on someone else's.
What you get
- Feasibility assessment on your actual footage
- Model selection, training or fine-tuning on your data
- The pipeline, hitting frame rate on your hardware
- Confidence thresholds tuned to what a miss costs you
- A human review queue for anything the model is unsure of
- Measured accuracy, on your footage, written down
Typical timeline: Six to twenty weeks.
What we have built with this
FireSmokeDetector
Computer vision fire and smoke detection.
Research
GENOME
A content codec built from nothing, targeting 100:1.
Research
Adtomaton
Brief in. Platform-ready ad out.
In Development
4 of our own systems use this discipline. Every one is ours, and the ones marked open source are yours to read. The whole product line →
Where this comes up
Build a computer vision system
You need to detect, count, read, track or identify something on camera, and the off-the-shelf products either do not fit or want your footage in their cloud.
Automate review of photos and video
Staff manually review images or footage — for damage, compliance, hazards or identification — and it does not scale with volume.
How we work
- We talk it through
- A conversation about the problem, not a sales pitch. If we are not the right people for the job we will say so and point you somewhere better. Free, and there is no obligation after it.
- You get a written quote
- A fixed price for a fixed scope, in plain English, with a delivery date. No hourly billing that quietly doubles. If the scope changes later, you approve the change and its cost before any work starts.
- We build it in stages
- You see working software every week or two, not a status report. That means you can course-correct early, while it is still cheap to change your mind.
- You take ownership
- At handover you receive the source code, the accounts, the domains and the documentation, all in your business name. Nothing stays locked to us. You could hire anyone else tomorrow and they could pick it up.
- We look after it
- An ongoing support agreement, or on-call as you need it. Your choice, and you can change or end it whenever you like.
Last reviewed · Site changelog