# Text-to-Image with Visual condition on CelebA-TV2I (test)

Source: [Text-to-Image with Visual condition on CelebA-TV2I (test)](<https://www.sota2.com/research/sota/text-to-image-with-visual-condition-on-celeba-tv2i-test>)

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

Data updated: 2026-05-08T08:07:17.783Z

Evaluation of Text-to-Image generation models conditioned on both text prompts and visual inputs, using the test split of the CelebA-TV2I dataset. Metrics include ID-Distance, Face Detection Rate, Condition Consistency, and General Fidelity.

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-to-Image with Visual condition |
| Dataset (including any stated split) | CelebA-TV2I (test) |
| Metric | ID-Distance |
| Unit | SCALAR |
| Direction | Lower is better |
| Results loaded | 5 of 5 |
| 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

- [ID-Distance](<https://www.sota2.com/research/sota/text-to-image-with-visual-condition-on-celeba-tv2i-test.md?metric=ID-Distance>): Lower is better; unit: SCALAR.

- [Face Detection Rate](<https://www.sota2.com/research/sota/text-to-image-with-visual-condition-on-celeba-tv2i-test.md?metric=Face%20Detection%20Rate>): Higher is better; unit: PERCENT.

- [Condition Consistency](<https://www.sota2.com/research/sota/text-to-image-with-visual-condition-on-celeba-tv2i-test.md?metric=Condition%20Consistency>): Higher is better; unit: PERCENT.

- [General Fidelity](<https://www.sota2.com/research/sota/text-to-image-with-visual-condition-on-celeba-tv2i-test.md?metric=General%20Fidelity>): Higher is better; unit: PERCENT.

## Recorded results

| Method | ID-Distance | Result date | Report paper |
| --- | --- | --- | --- |
| SD inpainting | 0.408 | Not recorded | 68fa75e11e209131cea6a68f |
| SOW | 0.771 | Not recorded | 68fa75e11e209131cea6a68f |
| ControlNet | 1.092 | Not recorded | 68fa75e11e209131cea6a68f |
| TI | 1.201 | Not recorded | 68fa75e11e209131cea6a68f |
| DreamBooth | 1.326 | Not recorded | 68fa75e11e209131cea6a68f |

### SD inpainting

ID-Distance: 0.408. Result ID: 69fd993b47f34ce924e07466.

[Reporting paper](<https://www.sota2.com/research/paper/68fa75e11e209131cea6a68f.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa6c391e209131cea32e4a.md>)

[Implementation](<https://github.com/CompVis/latent-diffusion>)

Recorded evaluation configuration: {"Time Cost":"5s","Stable Diffusion Version":"sd-v2-1-base"}. This record does not assert independent reproduction.

### SOW

ID-Distance: 0.771. Result ID: 69fd993b47f34ce924e07467.

[Reporting paper](<https://www.sota2.com/research/paper/68fa75e11e209131cea6a68f.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa75e11e209131cea6a68f.md>)

Recorded evaluation configuration: {"Time Cost":"5s","Stable Diffusion Version":"sd-v2-1-base"}. This record does not assert independent reproduction.

### ControlNet

ID-Distance: 1.092. Result ID: 69fd993b47f34ce924e07465.

[Reporting paper](<https://www.sota2.com/research/paper/68fa75e11e209131cea6a68f.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa71481e209131cea50695.md>)

Recorded evaluation configuration: {"Time Cost":"4s","Stable Diffusion Version":"sd-v2-1-base","Condition":"Canny edges"}. This record does not assert independent reproduction.

### TI

ID-Distance: 1.201. Result ID: 69fd993b47f34ce924e07463.

[Reporting paper](<https://www.sota2.com/research/paper/68fa75e11e209131cea6a68f.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa6d2d1e209131cea38190.md>)

Recorded evaluation configuration: {"Time Cost":"3025s","Stable Diffusion Version":"sd-v1-5"}. This record does not assert independent reproduction.

### DreamBooth

ID-Distance: 1.326. Result ID: 69fd993b47f34ce924e07464.

[Reporting paper](<https://www.sota2.com/research/paper/68fa75e11e209131cea6a68f.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa6ec31e209131cea41d81.md>)

[Implementation](<https://dreambooth.github.io/>)

Recorded evaluation configuration: {"Time Cost":"732s","Stable Diffusion Version":"sd-v1-5"}. This record does not assert independent reproduction.

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