| <div style="text-align: center; max-width: 650px; margin: 0 auto;"> | |
| <div style=" | |
| display: inline-flex; | |
| gap: 0.8rem; | |
| font-size: 1.75rem; | |
| justify-content: center; | |
| margin-bottom: 10px; | |
| "> | |
| <h1 style="font-weight: 900; align-items: center; margin-bottom: 7px; margin-top: 20px;"> | |
| Text-guided Background Generation for Salient Objects 🎨 | |
| </h1> | |
| </div> | |
| <div> | |
| <p style="align-items: center; margin-bottom: 7px;"> | |
| Create a new background for an image with a visible salient object using a text prompt. This space demos the "object expansion" issue when using inpainting models for background generation and how it can be fixed using <a href="https://huggingface.co/yahoo-inc/photo-background-generation">photo-background-generation</a> model. We use <a href="https://pypi.org/project/transparent-background/">transparent-background</a> to obtain the foreground mask. The research paper of this work: <a href="https://arxiv.org/abs/2404.10157">Arxiv</a> | |
| </p> | |
| </div> | |
| </div> | |