stable-dreamfusion
GET3D
stable-dreamfusion | GET3D | |
---|---|---|
41 | 17 | |
7,813 | 4,122 | |
- | 0.6% | |
7.2 | 4.2 | |
5 months ago | 6 months ago | |
Python | Python | |
Apache License 2.0 | GNU General Public License v3.0 or later |
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.
stable-dreamfusion
-
When are we getting stable diffusion for 3d models or 3d scenes?
Who is working on it? I've seen a few other models that do this, like Stable-Dreamfusion.
-
Is it possible for me to approximate a depth map from a generated image and make a 3D model?
I haven't tried Stable-DreamFusion, but it might be able to take an input image along with a prompt?
-
Meet ProlificDreamer: An AI Approach That Delivers High-Fidelity and Realistic 3D Content Using Variational Score Distillation (VSD)
similar to Magic3D and Dreamfusion / Stable-Dreamfusion, but this one looks a lot more vivid and detailed!
-
How would you all feel about 3D Stable Diffusion?
I've seen a few "text-to-3D" models that use Stable Diffusion. Zero-1-to-3 and Stable-DreamFusion appear to be capable of generating 3D models from text prompts.
-
Do any other software devs feel left behind by AI? I feel like I'm working on yesterday's tech
Ever heard of Stable Dreamfusion ? Open source text to 3D mesh model
-
Text-to-image-to-3D on 16GB GPU after stable-dreamfusion repo update
I followed the steps in this repo. They added my turtle as an example to the repo after their latest improvements. https://github.com/ashawkey/stable-dreamfusion
- I recreated Beat Saber in Unity only following Chat GPT. The code, the VFX and even the 3D models are made by an AI. Full video in the first comment.
-
ControlNet v1.1 has been released
There is an (independent?) implementation here, released last week to 0.1, but it already has a 100 issues.
-
Would it be possible for SD to make 3D deisgns that I can later 3D print?
But there's stable-dreamfusion, still not good but you could try it.
-
Game prototype using AI assisted graphics
Dreamfusion -- text-to-3d seems like it could be useful here: https://dreamfusion3d.github.io/ (once successfully opensourced, see https://github.com/ashawkey/stable-dreamfusion)
Rigging also looks like it could have a decent AI/DNN solution: https://arxiv.org/pdf/2005.00559.pdf
GET3D
-
Egregoria is a city simulation with high granularity
Check out: https://github.com/nv-tlabs/GET3D
License looks open. Very demanding hardware requirements (8 A100s).
- Tell HN: The next generation of videogames will be great with midjourney
-
Save money for figure by generate one
With CharTurner it might be possible to extrapolate out a 3D model 🤔 Nvidia was messing with it back in Fall of least year. https://github.com/nv-tlabs/GET3D
-
Creating 3D models using NVIDIA Get3D
Get 3D Code: https://github.com/nv-tlabs/GET3D
-
Ai generated 3d models?
GET3D
- Game development is about to go wild with AI
- Absolutely crazy - addon in blender to add textures to models with text prompts. Dream Textures (SD for Blender) - free addon. My god, it's accelerating at the speed of light.
-
Do you have any tips for converting 2D image to 3D, have you tried asking SD to do various different angles of the same image?
You are going to have a hard time from an SD image as you can't reliably get other angles to build a model, Nvidia's Get3D can try from a single image but the results are basic from what I can make out but I have not run it myself yet : https://github.com/nv-tlabs/GET3D
-
RTX 4090 Launch Thread - Surprise Inside
Honestly: I'm looking into getting a 4090, as someone with a 3090. I really, really push my card doing AI assisted video sequences, machine learning, simulations in blender, production for The largest metaverse convention in the world , and being able to record and stream to AV1 on the fly would be icing on the cake. Though I wish it had 48gb of Vram for custom ML training datasets to use nvidia's nerfs and tlabs Get3d, 24 will do for now. It also pushes through vr pretty well when paired with pci-e 5. Or running renderdoc or anything else. I'm currently working with a small team for an exhibit at Raindance Immersive, and I know the 30+ minutes it takes to generate 30 seconds of footage right now would be easially cut in half for this new show.
-
This room is dynamically generated with GPT-3 creating its person, items and writings, while SD paints the pictures on the wall.
Do you have the resources to run https://github.com/nv-tlabs/GET3D to try to AI generate objects GPT-3 references?
What are some alternatives?
instant-ngp - Instant neural graphics primitives: lightning fast NeRF and more
dreamgaussian - Generative Gaussian Splatting for Efficient 3D Content Creation
hashcat - World's fastest and most advanced password recovery utility
zero123plus - Code repository for Zero123++: a Single Image to Consistent Multi-view Diffusion Base Model.
magic3d-pytorch - Implementation of Magic3D, Text to 3D content synthesis, in Pytorch
ComfyUI_Noise - 6 nodes for ComfyUI that allows for more control and flexibility over noise to do e.g. variations or "un-sampling"
stable-diffusion-webui - Stable Diffusion web UI
stable-diffusion-webui-depthmap-script - High Resolution Depth Maps for Stable Diffusion WebUI
inline_tweak - Tweak a Rust value directly in the source code
point-e - Point cloud diffusion for 3D model synthesis