Machi - Concept Arts


Generated some concept arts for the city of Lael.
Some lessons:
What I noticed was that concept arts Midjourney generates is beyond anything I can create on my own. I might as well use it to the advantage.
Side scrolling view looks great but isometric and side views also look fantastic. As a player, it may make more sense to prefer the perspective views even though things are not editable.
In some sense, having fully perspective pixel art games were very expensive. AI opens the possibility.
Noticed Midjourney is also able to re-texture a picture without losing much of the contours. See last 3 images.
Instead of creating semantic map on the fly, what would happen if we generated lots of these scenes and embedded the semantics. Then train the semantics gen AI. Finally use the semantic map to generate the visuals back.
Next area of research: See how we can map semantic voxel map to perspective scenes.
Can voxel maps be used as a control net to create structures?
Can we generate semantic voxel map from a picture?
Can editing that voxel map allow for remix?
The perspective scenes (like the ones in Final Fantasy 7) is such an amazing cinematic presentation. I would not like to lose it.
Given that these generative powers are available at 10 dollar a month. The real differentiator for successful project won’t be just the visual. However, the first runners will definitely have their edge.
Original Concept Used:
Generated:
Edit mode allows you to expand the canvas naturally.
Subscribe to my newsletter
Read articles from Sprited Dev directly inside your inbox. Subscribe to the newsletter, and don't miss out.
Written by
