Extract all occurrences of timestamp in string

I have a string like this in Python:

string = "21042022_item1_21052022_item2_21062022_item3" 

How can I extract the dates from this string in a list?
Here’s my code:

import re

s = "21042022_item1_21052022_item2_21062022_item3"

print(re.findall('\d+',s))

but the result is not what I’m after:

['21042022', '1', '21052022', '2', '21062022', '3']

I don’t want the 1, 2 and 3.
Please help

>Solution :

This is a simple case, so technically it could be achievable without the re module also:

>>> string = "21042022_item1_21052022_item2_21062022_item3" 
>>> [x for x in string.split('_') if x.isnumeric()]
['21042022', '21052022', '21062022']

In case you’d like to work with date objects instead of str:

>>> from datetime import datetime
>>> [datetime.strptime(x, '%d%m%Y').date() for x in string.split('_') if x.isnumeric()]
[datetime.date(2022, 4, 21), datetime.date(2022, 5, 21), datetime.date(2022, 6, 21)]