I was really curious about the thought process here, but now I understand slightly more lol
A covering of tires could well be calculated to break up the infrared signature of these aircraft, to confuse cruise missiles using image matching for targeting. This technique is also frequently referred to as DSMAC (Digital Scene Matching Area Correlator) or ATR (Automated Target Recognition) when used in cruise missiles. As we have previously discussed, using DSMAC/ATR would provide land attack versions of Ukraine’s homegrown Neptune missiles with a significant advantage, making them largely immune to electronic warfare jamming. At the same time, their approach to the target would not involve any telltale radio-frequency emissions, thanks to the passive nature of the targeting.
Sounds like Ukraine needs to include some pictures of planes with tires on top in their training dataset.
Honestly I wonder if they could just train another, smaller model on widely-available aerial pictures of tires and develop a classical algorithm that steers towards things that look like a pile of tires on top of a plane shape. This is pretty shitty as adversarial patterns go.
Instructions unclear, redirecting to gokart arena!
This is absolutely GOLD, Jerry!
GOLD!
“he thinks everything that’s round is funny”
So hear me out you guys, what if the tires didn't have air in the, but... milk?