# Dense Video Object Captioning on VidSTG (test)

Source: [Dense Video Object Captioning on VidSTG (test)](<https://www.sota2.com/research/sota/dense-video-object-captioning-on-vidstg-test>)

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

Data updated: 2026-02-26T05:38:56.124Z

Evaluation of Dense Video Object Captioning performance on the test split of the VidSTG dataset, measured using metrics such as CapA, CHOTA, DetA, AssA, and AP\_M.

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 | Dense Video Object Captioning |
| Dataset (including any stated split) | VidSTG (test) |
| Metric | CapA |
| Unit | PERCENT |
| Direction | Higher 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

- [CapA](<https://www.sota2.com/research/sota/dense-video-object-captioning-on-vidstg-test.md?metric=CapA>): Higher is better; unit: PERCENT.

- [CHOTA](<https://www.sota2.com/research/sota/dense-video-object-captioning-on-vidstg-test.md?metric=CHOTA>): Higher is better; unit: PERCENT.

- [DetA](<https://www.sota2.com/research/sota/dense-video-object-captioning-on-vidstg-test.md?metric=DetA>): Higher is better; unit: PERCENT.

- [AssA](<https://www.sota2.com/research/sota/dense-video-object-captioning-on-vidstg-test.md?metric=AssA>): Higher is better; unit: PERCENT.

- [AP\_M](<https://www.sota2.com/research/sota/dense-video-object-captioning-on-vidstg-test.md?metric=AP_M>): Higher is better; unit: PERCENT.

## Recorded results

| Method | CapA | Result date | Report paper |
| --- | --- | --- | --- |
| OW-VISCapTor+CAROQ \[18\] (online) | 43.9 | Not recorded | 68fa77b71e209131cea74ef2 |
| OW-VISCapTor+DEVA \[7\] (online) | 40.1 | Not recorded | 68fa77b71e209131cea74ef2 |
| DVOC-DS (joint training) \[6\] | 36.8 | Not recorded | 68fa77b71e209131cea74ef2 |
| Mask2Former \[13\]+DEVA \[7\]+ BLIP2 \[11\] | 34 | Not recorded | 68fa77b71e209131cea74ef2 |
| DVOC-DS (disjoint training) \[6\] | 10 | Not recorded | 68fa77b71e209131cea74ef2 |

### OW-VISCapTor+CAROQ \[18\] (online)

CapA: 43.9. Result ID: 699560b23badfaffe653130d.

[Reporting paper](<https://www.sota2.com/research/paper/68fa77b71e209131cea74ef2.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa77b71e209131cea74ef2.md>)

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

Recorded evaluation configuration: {"Mode":"online"}. This record does not assert independent reproduction.

### OW-VISCapTor+DEVA \[7\] (online)

CapA: 40.1. Result ID: 699560b23badfaffe653130e.

[Reporting paper](<https://www.sota2.com/research/paper/68fa77b71e209131cea74ef2.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa77b71e209131cea74ef2.md>)

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

Recorded evaluation configuration: {"Mode":"online"}. This record does not assert independent reproduction.

### DVOC-DS (joint training) \[6\]

CapA: 36.8. Result ID: 699560b23badfaffe653130a.

[Reporting paper](<https://www.sota2.com/research/paper/68fa77b71e209131cea74ef2.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa75311e209131cea664d4.md>)

Recorded evaluation configuration: {"Mode":"offline"}. This record does not assert independent reproduction.

### Mask2Former \[13\]+DEVA \[7\]+ BLIP2 \[11\]

CapA: 34. Result ID: 699560b23badfaffe653130c.

[Reporting paper](<https://www.sota2.com/research/paper/68fa77b71e209131cea74ef2.md>)

Recorded evaluation configuration: {"Mode":"online"}. This record does not assert independent reproduction.

### DVOC-DS (disjoint training) \[6\]

CapA: 10. Result ID: 699560b23badfaffe653130b.

[Reporting paper](<https://www.sota2.com/research/paper/68fa77b71e209131cea74ef2.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa75311e209131cea664d4.md>)

Recorded evaluation configuration: {"Mode":"offline"}. This record does not assert independent reproduction.

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