Hey
I currently use Skechup + Vray for interior renders, of door hardware and things like turnstiles / speed gates.
It’s safe to say I mainly do interior renders, but also, when I have time, I do the odd exterior render.
Currently I tend to create my own models for the things in focus e.g. high quality door hardware.
In the background, I tend to pad it out with models from Vray, or better yet, from Poliigon.
Nine times out of ten, I blur the background, to reduce the size of the model as a whole (lower quality models in background).
Normally, for things beyond sight, I use HDRs with the odd building model from vray (which are very helpful by the way).
That being said, I know it’s a complicated topic, but is there any inkling as to to very basic AI generated ‘background models’.
I’m thinking things like:
- European village X stories
- USA village X stories
- Interior commercial
- Interor home
Better yet, if you create a house, for the AI to populate ‘the village’, with models of a similar nature.
I see AI generated imagery is generally going really far now, personally I use lexica and photoshop for abstract backgrounds and the likes of. Really inspring.
Just wondering if this is something that could be possible in the near future?
I’m just about to work on a render of a country house. So far it’s look ok, but no HDR can hide the fact that it’s floating in space so to speak. Even with primitive background models, it makes it feel much more realistic.
I guess you could download presets ‘scenes’ of ‘the average village’, but tweaking per project would be nice.
Similar to Lexica, crowd sourcing the results of AI generated bg models could be nice too. Needless to say, this could be paired with cloud computing to create the models in the first place, due to the complexity of the task.
I would imagine the same AI algorithem could be used to create variants of chairs, tables and everything else in between… not a small task, but certainly could fit within the community / 3D warehouse / AIHouse!?
Best
T
