I don't think the reason can be easily worked out. The one-shot learning of background introducing error in an object is purely speculative reasoning. On top of that, in a real-world situation, the background never really stays static.
I tried to give them the benefit of doubt when they got mixed demo results in the Linley demo a few months back.
Not expecting iPhone quality presentation as someone is alluding to which is irrelevant in the discussion as I don't care about aesthetics.
Demo should at least work to its designed purpose. And the same inaccuracy in identifying the object happened again.
Add to My Watchlist
What is My Watchlist?
 (20min delay)
|
|||||
|
Last
22.5¢ |
Change
0.010(4.65%) |
Mkt cap ! $455.7M | |||
| Open | High | Low | Value | Volume |
| 21.5¢ | 22.5¢ | 21.0¢ | $3.057M | 13.97M |
Buyers (Bids)
| No. | Vol. | Price($) |
|---|---|---|
| 4 | 140842 | 22.0¢ |
Sellers (Offers)
| Price($) | Vol. | No. |
|---|---|---|
| 22.5¢ | 828616 | 25 |
View Market Depth
| Last trade - 16.10pm 18/07/2025 (20 minute delay) ? |
| BRN (ASX) Chart |