# Multimodal Oil Painting Generation on DiffusionDB stylized

Source: [Multimodal Oil Painting Generation on DiffusionDB stylized](<https://www.sota2.com/research/sota/multimodal-oil-painting-generation-on-diffusiondb-stylized>)

Collection index: [llms.txt](<https://www.sota2.com/research/llms.txt>)

Data updated: 2026-02-26T05:46:01.977Z

Evaluation of multimodal generation models focusing on producing oil paintings, using the DiffusionDB stylized dataset, assessed via CLIP similarity (Text/Image), Gram matrix similarity, FID, and Aesthetic scores.

A research result applies to its stated dataset, split, metric and evaluation configuration. Compare like-for-like settings, preserve the metric's unit and direction, and distinguish paper-reported results from independent reproduction. The best result recorded here is not a claim of exhaustive, current global coverage.

| Scope | Recorded value |
| --- | --- |
| Task | Multimodal Oil Painting Generation |
| Dataset (including any stated split) | DiffusionDB stylized |
| Metric | Gram Similarity |
| Unit | SCALAR |
| Direction | Higher is better |
| Results loaded | 8 of 8 |
| Coverage | All results currently returned for this benchmark |

Evaluation settings may differ between results. The original papers and recorded configurations are needed to establish comparability. A shared metric name alone does not establish an identical protocol.

## Metrics

- [Gram Similarity](<https://www.sota2.com/research/sota/multimodal-oil-painting-generation-on-diffusiondb-stylized.md?metric=Gram%20Similarity>): Higher is better; unit: SCALAR.

- [CLIP Similarity (Text)](<https://www.sota2.com/research/sota/multimodal-oil-painting-generation-on-diffusiondb-stylized.md?metric=CLIP%20Similarity%20(Text)>): Higher is better; unit: SCALAR.

- [CLIP Similarity (Image)](<https://www.sota2.com/research/sota/multimodal-oil-painting-generation-on-diffusiondb-stylized.md?metric=CLIP%20Similarity%20(Image)>): Higher is better; unit: SCALAR.

- [FID](<https://www.sota2.com/research/sota/multimodal-oil-painting-generation-on-diffusiondb-stylized.md?metric=FID>): Lower is better; unit: SCALAR.

- [Aesthetic Score](<https://www.sota2.com/research/sota/multimodal-oil-painting-generation-on-diffusiondb-stylized.md?metric=Aesthetic%20Score>): Higher is better; unit: SCALAR.

## Recorded results

| Method | Gram Similarity | Result date | Report paper |
| --- | --- | --- | --- |
| PaintFlow | 0.619 | Not recorded | 6938fe876cfd488761a53c88 |
| PaintFlow w/o semantic | 0.611 | Not recorded | 6938fe876cfd488761a53c88 |
| PaintFlow w/o text | 0.605 | Not recorded | 6938fe876cfd488761a53c88 |
| MagicQuill | 0.585 | Not recorded | 6938fe876cfd488761a53c88 |
| IP-Adapter+SD-ControlNet-Inpaint | 0.568 | Not recorded | 6938fe876cfd488761a53c88 |
| IP-Adapter+SD-ControlNet-Scribble | 0.431 | Not recorded | 6938fe876cfd488761a53c88 |
| PaintFlow w/o style | 0.42 | Not recorded | 6938fe876cfd488761a53c88 |
| Paint-by-Example | 0.417 | Not recorded | 6938fe876cfd488761a53c88 |

### PaintFlow

Gram Similarity: 0.619. Result ID: 69986e745698665ec4ce5bf0.

[Reporting paper](<https://www.sota2.com/research/paper/6938fe876cfd488761a53c88.md>)

[Method origin paper](<https://www.sota2.com/research/paper/6938fe876cfd488761a53c88.md>)

Recorded evaluation configuration: {"Target Alignment - Text":true,"Target Alignment - Semantics":true,"Target Alignment - Style":true}. This record does not assert independent reproduction.

### PaintFlow w/o semantic

Gram Similarity: 0.611. Result ID: 69986e745698665ec4ce5bf2.

[Reporting paper](<https://www.sota2.com/research/paper/6938fe876cfd488761a53c88.md>)

[Method origin paper](<https://www.sota2.com/research/paper/6938fe876cfd488761a53c88.md>)

Recorded evaluation configuration: {"Target Alignment - Text":true,"Target Alignment - Semantics":false,"Target Alignment - Style":true}. This record does not assert independent reproduction.

### PaintFlow w/o text

Gram Similarity: 0.605. Result ID: 69986e745698665ec4ce5bf1.

[Reporting paper](<https://www.sota2.com/research/paper/6938fe876cfd488761a53c88.md>)

[Method origin paper](<https://www.sota2.com/research/paper/6938fe876cfd488761a53c88.md>)

Recorded evaluation configuration: {"Target Alignment - Text":false,"Target Alignment - Semantics":true,"Target Alignment - Style":true}. This record does not assert independent reproduction.

### MagicQuill

Gram Similarity: 0.585. Result ID: 69986e745698665ec4ce5bef.

[Reporting paper](<https://www.sota2.com/research/paper/6938fe876cfd488761a53c88.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa77981e209131cea743e1.md>)

[Implementation](<https://magic-quill.github.io/>)

Recorded evaluation configuration: {"Target Alignment - Text":true,"Target Alignment - Semantics":false,"Target Alignment - Style":true}. This record does not assert independent reproduction.

### IP-Adapter+SD-ControlNet-Inpaint

Gram Similarity: 0.568. Result ID: 69986e745698665ec4ce5bec.

[Reporting paper](<https://www.sota2.com/research/paper/6938fe876cfd488761a53c88.md>)

Recorded evaluation configuration: {"Target Alignment - Text":true,"Target Alignment - Semantics":false,"Target Alignment - Style":true}. This record does not assert independent reproduction.

### IP-Adapter+SD-ControlNet-Scribble

Gram Similarity: 0.431. Result ID: 69986e745698665ec4ce5bed.

[Reporting paper](<https://www.sota2.com/research/paper/6938fe876cfd488761a53c88.md>)

Recorded evaluation configuration: {"Target Alignment - Text":true,"Target Alignment - Semantics":true,"Target Alignment - Style":false}. This record does not assert independent reproduction.

### PaintFlow w/o style

Gram Similarity: 0.42. Result ID: 69986e745698665ec4ce5bf3.

[Reporting paper](<https://www.sota2.com/research/paper/6938fe876cfd488761a53c88.md>)

[Method origin paper](<https://www.sota2.com/research/paper/6938fe876cfd488761a53c88.md>)

Recorded evaluation configuration: {"Target Alignment - Text":true,"Target Alignment - Semantics":true,"Target Alignment - Style":false}. This record does not assert independent reproduction.

### Paint-by-Example

Gram Similarity: 0.417. Result ID: 69986e745698665ec4ce5bee.

[Reporting paper](<https://www.sota2.com/research/paper/6938fe876cfd488761a53c88.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa6e061e209131cea3d803.md>)

[Implementation](<https://github.com/Fantasy-Studio/Paint-by-Example>)

Recorded evaluation configuration: {"Target Alignment - Text":false,"Target Alignment - Semantics":true,"Target Alignment - Style":false}. This record does not assert independent reproduction.

Page 1 of 1. This is an index page, not the full collection.
