My stuff

  • My Workflows

  • Liked Workflows

  • Following Workflows

Go to OpenArt main site
Upload workflow

[AVD] Text (or Image) to 3D (SDXL Lightning+TripoSR)

5.0

0 reviews
14
5.7K
1.2K
2
Description

Workflow for Advanced Visual Design class.


This is a quick and easy workflow utilizing the TripoSR model, which takes an image and converts it into a 3D model (OBJ).  


    • There is a switch in the middle of the workflow that lets you switch between using an image as the input or a text to image created image as the input.  


    • Again, for speed and quality, we are using SDXL Lightning model in the Efficient Loader to generate the image from text.  


    • One important step is the background remover, which creates a mask that removes the background from the subject of the image, so best results will be from images where there is good contrast between the foreground and background in the image.  


    • In the final node, you can rotate around the 3D model that has been generated to view it from all sides.  


    • Note: You will need to add the TripoSR nodes from ComfyUI Manager, and will need this TripoSR model https://huggingface.co/stabilityai/TripoSR/tree/main to make this work. Don't forget to rename the 'model.ckpt' to TripoSR.ckpt to make sure you know which model it is.
Discussion

(No comments yet)

Loading...

Author

5
5.1K
47
29.3K

No reviews yet

  • - latest (a year ago)

Primitive Nodes (10)

Note (6)

Primitive string multiline [Crystools] (1)

Reroute (2)

Switch image [Crystools] (1)

Custom Nodes (9)

ComfyUI

  • - LoadImage (1)

  • - PreviewImage (1)

ComfyUI-Flowty-TripoSR

  • - TripoSRViewer (1)

  • - TripoSRSampler (1)

  • - TripoSRModelLoader (1)

  • - Efficient Loader (1)

  • - KSampler (Efficient) (1)

  • - Image Remove Background Rembg (mtb) (1)

  • - Image Resize (1)

Checkpoints (0)

LoRAs (0)