# Audio-Visual Segmentation on VPO-MSMI 1.0 (test)

Source: [Audio-Visual Segmentation on VPO-MSMI 1.0 (test)](<https://www.sota2.com/research/sota/audio-visual-segmentation-on-vpo-msmi-1-0-test>)

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

Data updated: 2026-02-26T05:47:32.061Z

Evaluation of Audio-Visual Segmentation performance on the test split of the VPO-MSMI 1.0 dataset, reporting Jaccard (J), F-measure (FB), and combined J&FB 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 | Audio-Visual Segmentation |
| Dataset (including any stated split) | VPO-MSMI 1.0 (test) |
| Metric | J & FB Score |
| Unit | PERCENT |
| 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

- [J & FB Score](<https://www.sota2.com/research/sota/audio-visual-segmentation-on-vpo-msmi-1-0-test.md?metric=J%20%26%20FB%20Score>): Higher is better; unit: PERCENT.

- [Jaccard (J)](<https://www.sota2.com/research/sota/audio-visual-segmentation-on-vpo-msmi-1-0-test.md?metric=Jaccard%20(J)>): Higher is better; unit: PERCENT.

- [F-measure (FB)](<https://www.sota2.com/research/sota/audio-visual-segmentation-on-vpo-msmi-1-0-test.md?metric=F-measure%20(FB)>): Higher is better; unit: PERCENT.

## Recorded results

| Method | J & FB Score | Result date | Report paper |
| --- | --- | --- | --- |
| DDESeg | 68.39 | Not recorded | 68fa77751e209131cea733c3 |
| CPM\* | 68.07 | Not recorded | 68fa77751e209131cea733c3 |
| DDESeg | 67.9 | Not recorded | 68fa77751e209131cea733c3 |
| AVSStone | 58.76 | Not recorded | 68fa77751e209131cea733c3 |
| BiasAVS | 57.94 | Not recorded | 68fa77751e209131cea733c3 |
| CAVP | 56.48 | Not recorded | 68fa77751e209131cea733c3 |
| AVSegFormer | 49.93 | Not recorded | 68fa77751e209131cea733c3 |
| TPAVI | 43.19 | Not recorded | 68fa77751e209131cea733c3 |

### DDESeg

J & FB Score: 68.39. Result ID: 69948bd43a6c3148491debbb.

[Reporting paper](<https://www.sota2.com/research/paper/68fa77751e209131cea733c3.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa77751e209131cea733c3.md>)

Recorded evaluation configuration: {"Visual Backbone":"Trans.","Audio Backbone":"HTSAT","Trainable Params":"103.84M","Input Resolution":"224 x 224"}. This record does not assert independent reproduction.

### CPM\*

J & FB Score: 68.07. Result ID: 69948bd43a6c3148491debb9.

[Reporting paper](<https://www.sota2.com/research/paper/68fa77751e209131cea733c3.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa74f71e209131cea64ead.md>)

Recorded evaluation configuration: {"Visual Backbone":"ResNet50","Audio Backbone":"VGGish","Input Resolution":"Original"}. This record does not assert independent reproduction.

### DDESeg

J & FB Score: 67.9. Result ID: 69948bd43a6c3148491debba.

[Reporting paper](<https://www.sota2.com/research/paper/68fa77751e209131cea733c3.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa77751e209131cea733c3.md>)

Recorded evaluation configuration: {"Visual Backbone":"Trans.","Audio Backbone":"VGGish","Trainable Params":"103.84M","Input Resolution":"224 x 224"}. This record does not assert independent reproduction.

### AVSStone

J & FB Score: 58.76. Result ID: 69948bd43a6c3148491debb8.

[Reporting paper](<https://www.sota2.com/research/paper/68fa77751e209131cea733c3.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa74cc1e209131cea63eb5.md>)

Recorded evaluation configuration: {"Visual Backbone":"SwinBase","Audio Backbone":"VGGish","Trainable Params":"114.63M","Input Resolution":"224 x 224"}. This record does not assert independent reproduction.

### BiasAVS

J & FB Score: 57.94. Result ID: 69948bd43a6c3148491debb7.

[Reporting paper](<https://www.sota2.com/research/paper/68fa77751e209131cea733c3.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa74481e209131cea60be0.md>)

Recorded evaluation configuration: {"Visual Backbone":"SwinBase","Audio Backbone":"VGGish","Trainable Params":"107.12M","Input Resolution":"224 x 224"}. This record does not assert independent reproduction.

### CAVP

J & FB Score: 56.48. Result ID: 69948bd43a6c3148491debb6.

[Reporting paper](<https://www.sota2.com/research/paper/68fa77751e209131cea733c3.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa74831e209131cea62161.md>)

Recorded evaluation configuration: {"Visual Backbone":"ResNet50","Audio Backbone":"VGGish","Trainable Params":"119.79M","Input Resolution":"224 x 224"}. This record does not assert independent reproduction.

### AVSegFormer

J & FB Score: 49.93. Result ID: 69948bd43a6c3148491debb5.

[Reporting paper](<https://www.sota2.com/research/paper/68fa77751e209131cea733c3.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa71791e209131cea5174e.md>)

Recorded evaluation configuration: {"Visual Backbone":"PVT-V2-B5","Audio Backbone":"VGGish","Trainable Params":"186.05M","Input Resolution":"224 x 224"}. This record does not assert independent reproduction.

### TPAVI

J & FB Score: 43.19. Result ID: 69948bd43a6c3148491debb4.

[Reporting paper](<https://www.sota2.com/research/paper/68fa77751e209131cea733c3.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa6e8e1e209131cea408f0.md>)

Recorded evaluation configuration: {"Visual Backbone":"PVT-V2-B5","Audio Backbone":"VGGish","Trainable Params":"101.32M","Input Resolution":"224 x 224"}. This record does not assert independent reproduction.

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