# Visual Domain Classification on Visual Domain Decathlon Benchmark (VDDB) (test)

Source: [Visual Domain Classification on Visual Domain Decathlon Benchmark (VDDB) (test)](<https://www.sota2.com/research/sota/visual-domain-classification-on-visual-domain-decathlon-benchmark-vddb-test>)

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

Data updated: 2026-02-26T05:51:43.622Z

Evaluation of Visual Domain Classification performance on the test split of the Visual Domain Decathlon Benchmark (VDDB), reporting both accuracy metrics and model efficiency statistics (Parameters and FLOPs).

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 | Visual Domain Classification |
| Dataset (including any stated split) | Visual Domain Decathlon Benchmark (VDDB) (test) |
| Metric | Max Test Accuracy |
| Unit | PERCENT |
| Direction | Higher is better |
| Results loaded | 4 of 4 |
| 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

- [Max Test Accuracy](<https://www.sota2.com/research/sota/visual-domain-classification-on-visual-domain-decathlon-benchmark-vddb-test.md?metric=Max%20Test%20Accuracy>): Higher is better; unit: PERCENT.

- [Average Test Accuracy](<https://www.sota2.com/research/sota/visual-domain-classification-on-visual-domain-decathlon-benchmark-vddb-test.md?metric=Average%20Test%20Accuracy>): Higher is better; unit: PERCENT.

- [Std Dev Test Accuracy](<https://www.sota2.com/research/sota/visual-domain-classification-on-visual-domain-decathlon-benchmark-vddb-test.md?metric=Std%20Dev%20Test%20Accuracy>): Lower is better; unit: SCALAR.

- [Params/task (M)](<https://www.sota2.com/research/sota/visual-domain-classification-on-visual-domain-decathlon-benchmark-vddb-test.md?metric=Params%2Ftask%20(M)>): Lower is better; unit: SIZE.

- [FLOPs/task (x10^6)](<https://www.sota2.com/research/sota/visual-domain-classification-on-visual-domain-decathlon-benchmark-vddb-test.md?metric=FLOPs%2Ftask%20(x10%5E6)>): Lower is better; unit: OTHER.

## Recorded results

| Method | Max Test Accuracy | Result date | Report paper |
| --- | --- | --- | --- |
| μ2Net+ | 85.56 | Not recorded | 68fa6dd31e209131cea3c3a9 |
| muNet | 83.58 | Not recorded | 68fa6dd31e209131cea3c3a9 |
| Fine-tuning | 81.51 | Not recorded | 68fa6dd31e209131cea3c3a9 |
| Residual adapters | 80.67 | Not recorded | 68fa6dd31e209131cea3c3a9 |

### μ2Net+

Max Test Accuracy: 85.56. Result ID: 698fde973a6c314849fc6fce.

[Reporting paper](<https://www.sota2.com/research/paper/68fa6dd31e209131cea3c3a9.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa6dd31e209131cea3c3a9.md>)

[Implementation](<https://github.com/google-research/google-research/tree/master/muNet>)

Recorded evaluation configuration: Not recorded. This record does not assert independent reproduction.

### muNet

Max Test Accuracy: 83.58. Result ID: 698fde973a6c314849fc6fcd.

[Reporting paper](<https://www.sota2.com/research/paper/68fa6dd31e209131cea3c3a9.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa6e031e209131cea3d686.md>)

[Implementation](<https://github.com/google-research/google-research/tree/master/muNet>)

Recorded evaluation configuration: Not recorded. This record does not assert independent reproduction.

### Fine-tuning

Max Test Accuracy: 81.51. Result ID: 698fde973a6c314849fc6fcb.

[Reporting paper](<https://www.sota2.com/research/paper/68fa6dd31e209131cea3c3a9.md>)

Recorded evaluation configuration: {"frozen\_layers":"below 1st transformer layer"}. This record does not assert independent reproduction.

### Residual adapters

Max Test Accuracy: 80.67. Result ID: 698fde973a6c314849fc6fcc.

[Reporting paper](<https://www.sota2.com/research/paper/68fa6dd31e209131cea3c3a9.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa63d01e209131cea00e18.md>)

Recorded evaluation configuration: {"dimension":256}. This record does not assert independent reproduction.

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