Challenge: Depth-to-Image Generation

Text-guided depth-to-image generation

Here’s another interesting task performed by diffusion models. With text-guided depth-to-image generation, textual descriptions guide the generation of images from depth information. This results in realistic and detailed images with the help of textual descriptions. We can also define negative prompts to prevent what should not be included in image creation.

The model we will be using is "stabilityai/stable-diffusion-2-depth". The prompt and negative prompt can be sent as parameters to prompt and negative_prompt, respectively, while defining pipe.

Complete the following steps:

1. Update the name of the images based on the name of the uploaded file.