there are techniques that can prevent this, but you pay a real price in overall performance, which may or may not be worth it.
of course it's harder to make physical objects work, but if somebody did manage to make a method for physical adversarial attacks that worked reliably, I would guess you'd run into the same problems. it's not at all clear that the endgame is that we can easily stop these attacks.
Clothes change much slower than software updates, and defending against a few adversarial examples is relatively easy, so this shouldn't be a problem.
I change my clothes at least once a day and add to my wardrobe on a ~weekly to ~monthly basis. Having been party to large system upgrades at even a reasonably competent company, I'd be shocked to find out that any surveillance system anywhere is matching that velocity.
So who says it has to be a few specific images? They could auto-generate a different selection to choose from for each purchaser and then print to order. Every individual article of clothing could be unique.
Some of these approaches are 7+ years old and still work.
As soon as they update the software, they run it over the stored video and investigate the differences. When they identify particular evasion techniques, they can flag up all instances and see what was done under their cover.
This is why I'm focusing my efforts on political solutions to protect our privacy, rather than technical solutions. That's not to say I don't think technical solutions are important, because they absolutely are, but even things like encryption could be made illegal politically if they want to. Thus politics is the most important arena for privacy advocates IMHO.
But it feels like it has gone the other way now, to the point that people dismiss potentially helpful technical solutions because "what really matters is politics" even though in reality you need both.
If the government bans privacy technologies then it's hard to use them, but if nobody develops or uses them to begin with then you don't have them either.
And for some people the government in question isn't ours, it's a foreign authoritarian government where "policy solutions" are just not going to happen and the only option available is technology. It's a lot easier to develop that technology in less authoritarian countries, which means we have a responsibility to do it, because if they have to do it the cost of a mistake is a lot higher.
So I really should re-phrase my position to clarify that I meant in the United States, but also even in the US I do think the technological solutions are important and should be developed. The important point I want to stress is that I don't think we should ever disregard the political war because of technological confidence in our ability to thwart the authoritarians. Both politics and technology are both critical pieces. If we lose on either front, we lose.
Like this: https://xkcd.com/1105/
There's also the non-technical approaches, like https://en.wikipedia.org/wiki/Anti-mask_law.