zero123
StableCascade
zero123 | StableCascade | |
---|---|---|
6 | 4 | |
2,503 | 6,340 | |
2.9% | 2.9% | |
6.9 | 7.2 | |
5 months ago | 2 months ago | |
Python | Jupyter Notebook | |
MIT License | MIT License |
Stars - the number of stars that a project has on GitHub. Growth - month over month growth in stars.
Activity is a relative number indicating how actively a project is being developed. Recent commits have higher weight than older ones.
For example, an activity of 9.0 indicates that a project is amongst the top 10% of the most actively developed projects that we are tracking.
zero123
-
Stable Cascade
Someone with resources will have to train Zero123 [1] with this backbone.
[1] https://zero123.cs.columbia.edu/
-
Stable Zero123: Quality 3D Object Generation from Single Images
This looks a fine-tune of the classic zero123 (https://github.com/cvlab-columbia/zero123) I’m excited to check out the quality improvements.
Though 3d model synthesis is one use case, I found the less advertised base reprojection model to be more useful for gamedev at the moment. You can generate a multiview spritesheet from an image, and it’s fast enough for synthesis during a gameplay session. I couldn’t get a good quality/time balance to do the same with the 3d models, and the lack of mesh rigging or animation combined with imperfections in a fully 3d model tends to break the suspension of disbelief compared to what players are used to. I’m this will change as the tech develops and we layer more AI on top (automatic animation synthesis is an active research area).
If you’re interested in this you might also want to check out deforum (https://github.com/deforum-art/deforum-stable-diffusion) which provides even more powerful camera controls on top of stable diffusion designed for full scenes rather than single objects.
-
Text-to-image-to-3D on 16GB GPU after stable-dreamfusion repo update
As described in the stable-dreamfusion repo for the image to 3D using the zero123 model (you can read more about that in their repo here: https://github.com/cvlab-columbia/zero123) I used the 105000 checkpoint of zero123. It took about an hour to go through their initial NeRF generation and cleanup steps to get the model output.
-
NVIDIA presents GeNVS: Generative Novel View Synthesis with 3D-Aware Diffusion Models
Until then https://github.com/cvlab-columbia/zero123 was kinda okay, but practical results often left to be desired, from the imprecision of the view angles to the at times fanciful re-imaginations of the source object.
-
Zero-1-to-3: Zero-shot One Image to 3D Object
For anyone else who tried to download the weights and got Google Drive throwing a quota error at you, they're working on it: https://github.com/cvlab-columbia/zero123/issues/2
StableCascade
-
Stable Cascade Prompt Following Is Amazing — This Model Has Huge Potential — High Resolutions Uses Lesser VRAM
Stability AI recently released Stable Cascade. You can read here : https://github.com/Stability-AI/StableCascade
-
Stable Cascade VS StableCascade - a user suggested alternative
2 projects | 13 Feb 2024
-
Stable Cascade
I see in the commits that the license was changed from MIT to their own custom one: https://github.com/Stability-AI/StableCascade/commit/209a526...
Is it legal to use an older snapshot before the license was changed in accordance with the previous MIT license?
What are some alternatives?
stable-diffusion-webui-forge
stable-dreamfusion - Text-to-3D & Image-to-3D & Mesh Exportation with NeRF + Diffusion.
Fooocus - Focus on prompting and generating
ComfyUI-DiffusersStableCascade - Simple inference with StableCascade using diffusers in ComfyUI
ComfyUI_examples - Examples of ComfyUI workflows
genvs
StableCascade - Official Code for Stable Cascade
deforum-stable-diffusion
sd-webui-lobe-theme - 🅰️ Lobe theme - The modern theme for stable diffusion webui, exquisite interface design, highly customizable UI, and efficiency boosting features.