|
--- |
|
title: Stable Diffusion Texual Inversion Demo |
|
emoji: 📈 |
|
colorFrom: blue |
|
colorTo: purple |
|
sdk: gradio |
|
sdk_version: 4.41.0 |
|
app_file: app.py |
|
pinned: false |
|
license: mit |
|
--- |
|
|
|
Check out the configuration reference at https://huggingface.co/docs/hub/spaces-config-reference |
|
|
|
- The app takes a prompt and number of inference steps as input |
|
- It provides a dropdown for 5 of the styles mentioned. This is a multi-select dropdown. |
|
1. [birb-style](https://huggingface.co/sd-concepts-library/birb-style) |
|
2. [cute-game-style](https://huggingface.co/sd-concepts-library/cute-game-style) |
|
3. [depthmap](https://huggingface.co/sd-concepts-library/depthmap) |
|
4. [line-art](https://huggingface.co/sd-concepts-library/line-art) |
|
5. [low-poly-hd-logos-icons](https://huggingface.co/sd-concepts-library/low-poly-hd-logos-icons) |
|
|
|
- The app provides an option of target size for generated images. There are 3 options |
|
1. 512x512 - best quality but very slow |
|
2. 256x256 - not good quality but still slow |
|
3. 128x128 - poor quality but faster |
|
- The app provides option to select the loss to be applied - Hue loss or Text-Image similarity loss. If Text-Image Similarity loss is selected, then the app takes the additional guidance prompt as input as well. |
|
|
|
|
