# Text-guided portrait sketch manipulation on CelebA-HQ Region of Interest

Source: [Text-guided portrait sketch manipulation on CelebA-HQ Region of Interest](<https://www.sota2.com/research/sota/text-guided-portrait-sketch-manipulation-on-celeba-hq-region-of-interest>)

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

Data updated: 2026-04-29T07:31:47.103Z

Evaluation of models performing text-guided portrait sketch manipulation using the Region of Interest subset of the CelebA-HQ dataset, typically measured using accuracy and preference 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 | Text-guided portrait sketch manipulation |
| Dataset (including any stated split) | CelebA-HQ Region of Interest |
| Metric | Accuracy |
| Unit | PERCENT |
| Direction | Higher is better |
| Results loaded | 3 of 3 |
| 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

- [Accuracy](<https://www.sota2.com/research/sota/text-guided-portrait-sketch-manipulation-on-celeba-hq-region-of-interest.md?metric=Accuracy>): Higher is better; unit: PERCENT.

- [Preference Rate](<https://www.sota2.com/research/sota/text-guided-portrait-sketch-manipulation-on-celeba-hq-region-of-interest.md?metric=Preference%20Rate>): Higher is better; unit: PERCENT.

## Recorded results

| Method | Accuracy | Result date | Report paper |
| --- | --- | --- | --- |
| PortraVec | 80.8 | Not recorded | 68fa75141e209131cea65969 |
| VectorFusion | 70.6 | Not recorded | 68fa75141e209131cea65969 |
| CLIPVG | 63.1 | Not recorded | 68fa75141e209131cea65969 |

### PortraVec

Accuracy: 80.8. Result ID: 69f1b2f347f34ce924d7d0d2.

[Reporting paper](<https://www.sota2.com/research/paper/68fa75141e209131cea65969.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa75141e209131cea65969.md>)

Recorded evaluation configuration: {"Resolution":"224x224","Number of Paths":"120"}. This record does not assert independent reproduction.

### VectorFusion

Accuracy: 70.6. Result ID: 69f1b2f347f34ce924d7d0d1.

[Reporting paper](<https://www.sota2.com/research/paper/68fa75141e209131cea65969.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa6df91e209131cea3d170.md>)

Recorded evaluation configuration: {"Resolution":"224x224","Number of Paths":"120"}. This record does not assert independent reproduction.

### CLIPVG

Accuracy: 63.1. Result ID: 69f1b2f347f34ce924d7d0d0.

[Reporting paper](<https://www.sota2.com/research/paper/68fa75141e209131cea65969.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa6f4a1e209131cea44d28.md>)

Recorded evaluation configuration: {"Resolution":"224x224","Number of Paths":"120"}. This record does not assert independent reproduction.

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