SD-$π$XL: Generating Low-Resolution Quantized Imagery via Score Distillation
Fuente:
arXiv
Enregistré dans:
| Auteurs principaux: | , |
|---|---|
| Format: | Preprint |
| Publié: |
2024
|
| Sujets: | |
| Accès en ligne: | |
| Tags: |
Ajouter un tag
Pas de tags, Soyez le premier à ajouter un tag!
|
| _version_ | 1866916428500697088 |
|---|---|
| author | Binninger, Alexandre Sorkine-Hornung, Olga |
| author_facet | Binninger, Alexandre Sorkine-Hornung, Olga |
| contents | Low-resolution quantized imagery, such as pixel art, is seeing a revival in modern applications ranging from video game graphics to digital design and fabrication, where creativity is often bound by a limited palette of elemental units. Despite their growing popularity, the automated generation of quantized images from raw inputs remains a significant challenge, often necessitating intensive manual input. We introduce SD-$π$XL, an approach for producing quantized images that employs score distillation sampling in conjunction with a differentiable image generator. Our method enables users to input a prompt and optionally an image for spatial conditioning, set any desired output size $H \times W$, and choose a palette of $n$ colors or elements. Each color corresponds to a distinct class for our generator, which operates on an $H \times W \times n$ tensor. We adopt a softmax approach, computing a convex sum of elements, thus rendering the process differentiable and amenable to backpropagation. We show that employing Gumbel-softmax reparameterization allows for crisp pixel art effects. Unique to our method is the ability to transform input images into low-resolution, quantized versions while retaining their key semantic features. Our experiments validate SD-$π$XL's performance in creating visually pleasing and faithful representations, consistently outperforming the current state-of-the-art. Furthermore, we showcase SD-$π$XL's practical utility in fabrication through its applications in interlocking brick mosaic, beading and embroidery design. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2410_06236 |
| institution | arXiv |
| publishDate | 2024 |
| record_format | arxiv |
| spellingShingle | SD-$π$XL: Generating Low-Resolution Quantized Imagery via Score Distillation Binninger, Alexandre Sorkine-Hornung, Olga Computer Vision and Pattern Recognition Graphics I.3.3 Low-resolution quantized imagery, such as pixel art, is seeing a revival in modern applications ranging from video game graphics to digital design and fabrication, where creativity is often bound by a limited palette of elemental units. Despite their growing popularity, the automated generation of quantized images from raw inputs remains a significant challenge, often necessitating intensive manual input. We introduce SD-$π$XL, an approach for producing quantized images that employs score distillation sampling in conjunction with a differentiable image generator. Our method enables users to input a prompt and optionally an image for spatial conditioning, set any desired output size $H \times W$, and choose a palette of $n$ colors or elements. Each color corresponds to a distinct class for our generator, which operates on an $H \times W \times n$ tensor. We adopt a softmax approach, computing a convex sum of elements, thus rendering the process differentiable and amenable to backpropagation. We show that employing Gumbel-softmax reparameterization allows for crisp pixel art effects. Unique to our method is the ability to transform input images into low-resolution, quantized versions while retaining their key semantic features. Our experiments validate SD-$π$XL's performance in creating visually pleasing and faithful representations, consistently outperforming the current state-of-the-art. Furthermore, we showcase SD-$π$XL's practical utility in fabrication through its applications in interlocking brick mosaic, beading and embroidery design. |
| title | SD-$π$XL: Generating Low-Resolution Quantized Imagery via Score Distillation |
| topic | Computer Vision and Pattern Recognition Graphics I.3.3 |
| url | https://arxiv.org/abs/2410.06236 |