# Semantic Image Synthesis on DeepFashion

Source: [Semantic Image Synthesis on DeepFashion](<https://www.sota2.com/research/sota/semantic-image-synthesis-on-deepfashion>)

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

Data updated: 2026-02-27T06:40:31.458Z

Evaluation of models performing semantic image synthesis on the DeepFashion dataset, typically assessed using metrics covering semantic accuracy (mIoU, Accuracy) and perceptual quality (FID, LPIPS).

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 | Semantic Image Synthesis |
| Dataset (including any stated split) | DeepFashion |
| Metric | Params (M) |
| Unit | SIZE |
| Direction | Lower is better |
| Results loaded | 16 of 16 |
| 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

- [Params (M)](<https://www.sota2.com/research/sota/semantic-image-synthesis-on-deepfashion.md?metric=Params%20(M)>): Lower is better; unit: SIZE.

- [mIoU](<https://www.sota2.com/research/sota/semantic-image-synthesis-on-deepfashion.md?metric=mIoU>): Higher is better; unit: PERCENT.

- [Accuracy](<https://www.sota2.com/research/sota/semantic-image-synthesis-on-deepfashion.md?metric=Accuracy>): Higher is better; unit: PERCENT.

- [FID](<https://www.sota2.com/research/sota/semantic-image-synthesis-on-deepfashion.md?metric=FID>): Lower is better; unit: SCALAR.

- [LPIPS](<https://www.sota2.com/research/sota/semantic-image-synthesis-on-deepfashion.md?metric=LPIPS>): Lower is better; unit: SCALAR.

- [FLOPs (G)](<https://www.sota2.com/research/sota/semantic-image-synthesis-on-deepfashion.md?metric=FLOPs%20(G)>): Lower is better; unit: SCALAR.

- [Latency (s)](<https://www.sota2.com/research/sota/semantic-image-synthesis-on-deepfashion.md?metric=Latency%20(s)>): Lower is better; unit: TIME.

## Recorded results

| Method | Params (M) | Result date | Report paper |
| --- | --- | --- | --- |
| DSCGAN | 54 | Not recorded | 68fa69801e209131cea22806 |
| BicycleGAN | 54.8 | Not recorded | 68fa69801e209131cea22806 |
| CLADE | 66.98 | Not recorded | 68fa69801e209131cea22806 |
| INADE | 84.63 | Not recorded | 68fa69801e209131cea22806 |
| SPADE | 92.21 | Not recorded | 68fa69801e209131cea22806 |
| GroupDNet | 96.32 | Not recorded | 68fa69801e209131cea22806 |
| pix2pixHD | 182.44 | Not recorded | 68fa69801e209131cea22806 |
| SEAN | 223.23 | Not recorded | 68fa69801e209131cea22806 |

### DSCGAN

Params (M): 54. Result ID: 69a13c4435dfcabb00e5abe3.

[Reporting paper](<https://www.sota2.com/research/paper/68fa69801e209131cea22806.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa654b1e209131cea095b7.md>)

Recorded evaluation configuration: {"Evaluation Device":"Titan XP","Inference Scope":"generator networks only"}. This record does not assert independent reproduction.

### BicycleGAN

Params (M): 54.8. Result ID: 69a13c4435dfcabb00e5abe2.

[Reporting paper](<https://www.sota2.com/research/paper/68fa69801e209131cea22806.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa64f61e209131cea07624.md>)

[Implementation](<https://github.com/junyanz/BicycleGAN>)

Recorded evaluation configuration: {"Evaluation Device":"Titan XP","Inference Scope":"generator networks only"}. This record does not assert independent reproduction.

### CLADE

Params (M): 66.98. Result ID: 69a13c4435dfcabb00e5abe0.

[Reporting paper](<https://www.sota2.com/research/paper/68fa69801e209131cea22806.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa69ea1e209131cea24e0c.md>)

Recorded evaluation configuration: {"Evaluation Device":"Titan XP","Inference Scope":"generator networks only"}. This record does not assert independent reproduction.

### INADE

Params (M): 84.63. Result ID: 69a13c4435dfcabb00e5abe5.

[Reporting paper](<https://www.sota2.com/research/paper/68fa69801e209131cea22806.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa69801e209131cea22806.md>)

[Implementation](<https://github.com/tzt101/INADE.git>)

Recorded evaluation configuration: {"Evaluation Device":"Titan XP","Inference Scope":"generator networks only"}. This record does not assert independent reproduction.

### SPADE

Params (M): 92.21. Result ID: 69a13c4435dfcabb00e5abde.

[Reporting paper](<https://www.sota2.com/research/paper/68fa69801e209131cea22806.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa669c1e209131cea11672.md>)

[Implementation](<https://github.com/NVlabs/SPADE>)

Recorded evaluation configuration: {"Evaluation Device":"Titan XP","Inference Scope":"generator networks only"}. This record does not assert independent reproduction.

### GroupDNet

Params (M): 96.32. Result ID: 69a13c4435dfcabb00e5abe4.

[Reporting paper](<https://www.sota2.com/research/paper/68fa69801e209131cea22806.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa67561e209131cea159e4.md>)

Recorded evaluation configuration: {"Evaluation Device":"Titan XP","Inference Scope":"generator networks only"}. This record does not assert independent reproduction.

### pix2pixHD

Params (M): 182.44. Result ID: 69a13c4435dfcabb00e5abdf.

[Reporting paper](<https://www.sota2.com/research/paper/68fa69801e209131cea22806.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa64c11e209131cea06591.md>)

Recorded evaluation configuration: {"Evaluation Device":"Titan XP","Inference Scope":"generator networks only"}. This record does not assert independent reproduction.

### SEAN

Params (M): 223.23. Result ID: 69a13c4435dfcabb00e5abe1.

[Reporting paper](<https://www.sota2.com/research/paper/68fa69801e209131cea22806.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa682c1e209131cea1a882.md>)

Recorded evaluation configuration: {"Evaluation Device":"Titan XP","Inference Scope":"generator networks only"}. This record does not assert independent reproduction.

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