04 / VISION SYSTEMS

Computer vision systems we've built.

Detection, tracking, and inspection that run on real cameras in real places, plus the datasets, dashboards, and pipelines around them. Computer vision is where NNCODE started.

LIVE DEMOS

Watch them work, or press a key and try.

Fig 1Inspection line
Watch, or press a button
line stopped

Computer vision

Cameras that detect, count, and flag defects on real production lines, in real time. It's where NNCODE started.

Fig 2Detection & tracking
Watch, or press a button
camera ready

Detection & tracking

Detect, track, and count on a live camera feed, with the model's output on screen.

Try it on your own photo

Show us a photo. We'll show you what vision could do.

Drop in a photo from your business, and a vision-language model suggests what cameras and vision models could do in a place like that.

Show us a photoA warehouse aisle, a conveyor, a production line, a shelf, a site. Anything from your business.

or drop it here
vision · gemini 3.8 flashREADY
  • Preparing your photo
  • Uploading
  • Finding objects
  • Writing ideas
Waiting for a photo0.0 s

Or try an example

Your photo is shrunk on your device (or on our server if your browser can't open it) and sent to Google Gemini to be analysed. It isn't stored. Please don't upload anything confidential.

Want to see how we build these? How we build vision systems

SEE IT RUN

A vision system, in motion.

ROBOTICS · COMPUTER VISION

Autonomous forklift

A forklift that takes work orders in plain English and moves pallets on its own, with or without a map, around people and whatever is in the way. Shown in a simulated warehouse, sped up 4 to 13 times.

01 / 08

Rack to rack, across two halls

Takes a pallet out of a rack, drives through the fire door and past people, and puts it into a rack in hall B.

HOW THEY'RE BUILT

From pixels to decisions.

Fig 3Vision pipeline
Watch, or press a button
pipeline idle · 5 stations

WHAT WE'VE BUILT

Platforms, systems, and pipelines.

WRITING

What we've learned building these.

Computer visionHow multi-object tracking keeps the same ID on the same person7 min readComputer visionBounding boxes vs segmentation masks: what masks actually add6 min readComputer visionHow OCR reads a shipping label, and how to catch its mistakes7 min readComputer visionFrom pixels to floor positions: making camera zones work in metres6 min read3D visionDepth from two cameras: how stereo vision measures distance6 min readComputer visionNon-maximum suppression: why your detector draws one box, not six6 min readComputer visionChoosing the confidence threshold: misses versus false alarms7 min readComputer visionLighting decides what your inspection camera can see7 min readPerformanceQuantization: making a vision model smaller and faster without breaking it7 min read
Looking for AI agents and integrations instead?

Let's talk

Got something you want to build?

Tell us about it. You talk directly with the person who builds it, from the first chat to everything after.

  • Reply within a day
  • Straight to the builders
  • No obligation