|
--- |
|
license: creativeml-openrail-m |
|
tags: |
|
- pytorch |
|
- diffusers |
|
- stable-diffusion |
|
- text-to-image |
|
- diffusion-models-class |
|
- dreambooth-hackathon |
|
- science |
|
widget: |
|
- text: Photorealistic trex dinosaur, (Jurassic Park) lush greenery, towering prehistoric |
|
ferns, dramatic lighting casting shadows, (intense expression), rugged texture |
|
of scales, ferocious stance, misty atmosphere, high-quality detail, 4K, (cinematic |
|
depth), vibrant colors of nature, embedding a sense of ancient history, (dynamic |
|
environment) teeming with life and excitement. |
|
--- |
|
|
|
# A DreamBooth model capable of spectacular/cinematic visuals of Tyrannosaurus rex (the most famous and feared of all prehistoric predators) dinosaur image generator. |
|
Trained by HugeFighter on the HugeFighter/dinosaurs dataset, This is a Stable Diffusion model fine-tuned on the dinosaur concept with DreamBooth. |
|
It can be used by modifying the `instance_prompt`: **a photo of a trex dinosaur** |
|
|
|
This model was created as part of the DreamBooth Hackathon 🔥. Visit the [organisation page](https://huggingface.co/dreambooth-hackathon) for instructions on how to take part! |
|
|
|
## Description |
|
|
|
|
|
This is a Stable Diffusion model fine-tuned on `dinosaur` images for the science theme. |
|
|
|
|
|
## Usage |
|
|
|
```python |
|
from diffusers import StableDiffusionPipeline |
|
|
|
pipeline = StableDiffusionPipeline.from_pretrained('HugeFighter/trex-dinosaur') |
|
image = pipeline().images[0] |
|
image |
|
``` |
|
|
|
### Dataset Summary: T.Rex generator |
|
This dataset includes 12 images of trex collected randomly from web. |
|
The dataset contains unique photographs of trex with different backgrounds. |
|
I won't recommend this using these images as your training data (small dataset), but you can still use it for sampling purpose. |
|
|
|
The dataset is intended to be used in the context of the Hugging Face Dream Booth hackathon, a competition that challenges participants to build innovative applications using the Hugging Face transformers library. The submission is for the category of animal. |
|
|
|
## Dataset Structure |
|
|
|
The complete dataset consists of 12 objects, representing 0.01GB of stored data across 12 rows in parquet format. |
|
|
|
# Examples of output images generated by pipeline |
|
prompt = "A photo of a trex dinosaur, wandering over Jurassic park with a big roar" |
|
 |
|
|
|
prompt = "Photorealistic trex dinosaur, (Jurassic Park) lush greenery, towering prehistoric ferns, dramatic lighting casting shadows, (intense expression), rugged texture of scales, ferocious stance, misty atmosphere, high-quality detail, 4K, (cinematic depth), vibrant colors of nature, embedding a sense of ancient history, (dynamic environment) teeming with life and excitement." |
|
 |
|
|
|
|
|
|
|
|
