Hey Thierry - that's what the capability is supposed to offer you. There are multiple ways: textual inversion is the easiest to experiment with. You can find some representative works and train it as a style. If that hold potential, you could look at some of the newer approaches: hypernetwork (I haven't done it) and DreamBooth.

For textual inversion styling, you can read my approach here: https://ericri.medium.com/you-the-multiverse-of-you-and-stable-diffusion-93c2b8b8b3f6

If you're doing this at home on a strong enough PC, you can setup Automatic1111 as your environment - info here: https://ericri.medium.com/starting-out-with-stable-diffusion-9d2f6b274d94

ALSO, there are sooooo many videos on YouTube that cover this same material. Some by techie nerds like me, some by digital artists.

With the training, you should be able to do something like:

Photo shoot of <fill in the blank>, <any keywords you like>, professional photo in the style of Thierry Van Biesen