StyleShot: A Snapshot on Any Style
About
In this paper, we show that, a good style representation is crucial and sufficient for generalized style transfer without test-time tuning. We achieve this through constructing a style-aware encoder and a well-organized style dataset called StyleGallery. With dedicated design for style learning, this style-aware encoder is trained to extract expressive style representation with decoupling training strategy, and StyleGallery enables the generalization ability. We further employ a content-fusion encoder to enhance image-driven style transfer. We highlight that, our approach, named StyleShot, is simple yet effective in mimicking various desired styles, i.e., 3D, flat, abstract or even fine-grained styles, without test-time tuning. Rigorous experiments validate that, StyleShot achieves superior performance across a wide range of styles compared to existing state-of-the-art methods. The project page is available at: https://styleshot.github.io/.
Related benchmarks
| Task | Dataset | Result | Rank | |
|---|---|---|---|---|
| Style Transfer | ArtFID Benchmark (test) | ArtFID15.928 | 45 | |
| Image Style Transfer | User Study | Overall Quality Score76.6 | 30 | |
| Color-constrained image generation | color-constrained image generation | Latency (s)44.91 | 13 | |
| Color-constrained image generation | Color-constrained generation benchmark | HistKL Divergence1.9 | 12 | |
| Text-to-image Style Transfer | WikiArt | Text Score0.274 | 11 | |
| Image Style Transfer | Style Transfer 750 images (test) | Style Score0.5198 | 10 | |
| Multi-style Image Transfer | MS-COCO (content) & WikiArt (style) Two-style setting Stable Diffusion v1.4 backbone (test) | ArtFID17.966 | 9 | |
| Textile pattern generation | CTP-HD (with Ground Truth) | FID39.76 | 9 | |
| Style Transfer | CIFAR-100 and InstaStyle (test) | Content Score26.9 | 9 | |
| Style Transfer | Style Transfer Evaluation Set (test) | Style Score63.42 | 8 |