Product decision to screen acknowledgement
Lab note, 30 Sep 2026: from product decision to the screen's render acknowledgement, median 0.35 s with one product, 0.51 s with two. Method and limits.
From the same test session as the reaction-time note: how many fires reached the screen, and how GoTrack behaved when shirts were being worn in front of the cameras.

All 25 fires of the session were sent to the screen, and 24 reported their render back. In the same session, 30 sightings fell in the range GoTrack treats as clothing being worn rather than held, and none of them reached the screen; the same shirts were shown when they were held up.
| What | Count |
|---|---|
| Fires sent to the screen | 25 of 25 |
| Fires that reported their render | 24 of 25 |
| Sightings kept off the screen as worn clothing | 30 |
| Different products in the session | 8 |
In a store the cameras also see shoppers and staff in their own clothes. A screen that reacted every time someone walked past wearing something similar would soon mean nothing, and could show a product nobody picked up.
GoTrack waits until a product is being shown, which is why these 30 sightings stayed off the screen while the same shirts were shown when someone held them up.
No. There is no face recognition, no face template, no biometric matching, and no shopper is recognised across visits.
Not by default: each analysed frame is discarded. If debug recording is switched on, the files stay on the store device and are deleted after 30 days by default. The only image that reaches the product catalogue is one an authorised user deliberately adds as a product photo.
Garments on racks and shelves, and shoe walls — products a shopper picks up to look at. A garment's size is not read from the image.
We will show you GoTrack on products like yours.
or write to info@gotonom.com