Great stuff, I recommend CSM platform as well, I have created a recent video regarding this. Also glad that I find your channel and subscribed to it since you are also at the beginning like me!
Now think about pixel art and 2D platformers. If you made a sprite have a front, back, left side, right side, top view, bottom view, trace all the features, insert internal skeleton frames, apply external anchor points on each perspective, inflate to connect skeletons and anchors, this should in theory create a perfect 3D pixel model for diorama environments? Now if we can get AI to accept a text description of how the sprite is to move in relation to the diorama, using pre-scripted action sequences (for example: "humanoid walking normally" whereby action-skeletons & anchors connect to object-skeletons & anchors), the AI need only understand its environment in the same way a self-driving vehicle navigates the road according to protocol, to make the animation sequence happen. This eliminates any artistic or technological need or skill to draw a person or place or thing and to make it do anything, if mere pixel art is as a child's building blocks, let the AI paint over it according to text input and manual touch-up, to pull the strings on the puppet for you? This can allow for an assembled library of objects with actions, digital toys for future playing with.
![Everyone Can Make 3D Animations Easily Now! [Monster Mash]](/img/1.gif)
Great stuff, I recommend CSM platform as well, I have created a recent video regarding this. Also glad that I find your channel and subscribed to it since you are also at the beginning like me!
Great vid ! Very helpful thank you
Glad it was helpful! ✌️
What about making it more realistic? The colour? The face?
I see you are also a 3ds max user! That`s neat!
I see you are also a 3ds max user! That's neat!
Now think about pixel art and 2D platformers. If you made a sprite have a front, back, left side, right side, top view, bottom view, trace all the features, insert internal skeleton frames, apply external anchor points on each perspective, inflate to connect skeletons and anchors, this should in theory create a perfect 3D pixel model for diorama environments? Now if we can get AI to accept a text description of how the sprite is to move in relation to the diorama, using pre-scripted action sequences (for example: "humanoid walking normally" whereby action-skeletons & anchors connect to object-skeletons & anchors), the AI need only understand its environment in the same way a self-driving vehicle navigates the road according to protocol, to make the animation sequence happen. This eliminates any artistic or technological need or skill to draw a person or place or thing and to make it do anything, if mere pixel art is as a child's building blocks, let the AI paint over it according to text input and manual touch-up, to pull the strings on the puppet for you? This can allow for an assembled library of objects with actions, digital toys for future playing with.
Tried this site to convert a 2d image of a sock into a 3D obj. It wasn’t the best result, none of the sick texture showed.
sepas jaleb bod
Danke für das Tutorial!
Bitte bitte! 🙏