# Open-Vocabulary Object Detection on OV-COCO-C (test)

Source: [Open-Vocabulary Object Detection on OV-COCO-C (test)](<https://www.sota2.com/research/sota/open-vocabulary-object-detection-on-ov-coco-c-test>)

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

Data updated: 2026-03-31T09:53:32.001Z

Evaluation of Open-Vocabulary Object Detection performance on the OV-COCO-C test set, reporting mAP50 and performance across various visual corruption types (e.g., Gaussian noise, fog, JPEG compression).

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 | Open-Vocabulary Object Detection |
| Dataset (including any stated split) | OV-COCO-C (test) |
| Metric | mAP@0.5 (Gauss) |
| Unit | PERCENT |
| Direction | Higher is better |
| Results loaded | 11 of 11 |
| 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

- [mAP@0.5 (Gauss)](<https://www.sota2.com/research/sota/open-vocabulary-object-detection-on-ov-coco-c-test.md?metric=mAP%400.5%20(Gauss)>): Higher is better; unit: PERCENT.

- [mAP@0.5 (Shot)](<https://www.sota2.com/research/sota/open-vocabulary-object-detection-on-ov-coco-c-test.md?metric=mAP%400.5%20(Shot)>): Higher is better; unit: PERCENT.

- [mAP@0.5 (Impulse)](<https://www.sota2.com/research/sota/open-vocabulary-object-detection-on-ov-coco-c-test.md?metric=mAP%400.5%20(Impulse)>): Higher is better; unit: PERCENT.

- [mAP@0.5 (Defocus)](<https://www.sota2.com/research/sota/open-vocabulary-object-detection-on-ov-coco-c-test.md?metric=mAP%400.5%20(Defocus)>): Higher is better; unit: PERCENT.

- [mAP@0.5 (Glass)](<https://www.sota2.com/research/sota/open-vocabulary-object-detection-on-ov-coco-c-test.md?metric=mAP%400.5%20(Glass)>): Higher is better; unit: PERCENT.

- [mAP@0.5 (Motion)](<https://www.sota2.com/research/sota/open-vocabulary-object-detection-on-ov-coco-c-test.md?metric=mAP%400.5%20(Motion)>): Higher is better; unit: PERCENT.

- [mAP@0.5 (Zoom)](<https://www.sota2.com/research/sota/open-vocabulary-object-detection-on-ov-coco-c-test.md?metric=mAP%400.5%20(Zoom)>): Higher is better; unit: PERCENT.

- [mAP@0.5 (Snow)](<https://www.sota2.com/research/sota/open-vocabulary-object-detection-on-ov-coco-c-test.md?metric=mAP%400.5%20(Snow)>): Higher is better; unit: PERCENT.

- [mAP@0.5 (Frost)](<https://www.sota2.com/research/sota/open-vocabulary-object-detection-on-ov-coco-c-test.md?metric=mAP%400.5%20(Frost)>): Higher is better; unit: PERCENT.

- [mAP@0.5 (Fog)](<https://www.sota2.com/research/sota/open-vocabulary-object-detection-on-ov-coco-c-test.md?metric=mAP%400.5%20(Fog)>): Higher is better; unit: PERCENT.

- [mAP@0.5 (Brightness)](<https://www.sota2.com/research/sota/open-vocabulary-object-detection-on-ov-coco-c-test.md?metric=mAP%400.5%20(Brightness)>): Higher is better; unit: PERCENT.

- [mAP@0.5 (Contrast)](<https://www.sota2.com/research/sota/open-vocabulary-object-detection-on-ov-coco-c-test.md?metric=mAP%400.5%20(Contrast)>): Higher is better; unit: PERCENT.

- [mAP@0.5 (Elastic)](<https://www.sota2.com/research/sota/open-vocabulary-object-detection-on-ov-coco-c-test.md?metric=mAP%400.5%20(Elastic)>): Higher is better; unit: PERCENT.

- [mAP@0.5 (Pixelate)](<https://www.sota2.com/research/sota/open-vocabulary-object-detection-on-ov-coco-c-test.md?metric=mAP%400.5%20(Pixelate)>): Higher is better; unit: PERCENT.

- [mAP@0.5 (JPEG)](<https://www.sota2.com/research/sota/open-vocabulary-object-detection-on-ov-coco-c-test.md?metric=mAP%400.5%20(JPEG)>): Higher is better; unit: PERCENT.

- [mAP@0.5 (Overall Average)](<https://www.sota2.com/research/sota/open-vocabulary-object-detection-on-ov-coco-c-test.md?metric=mAP%400.5%20(Overall%20Average)>): Higher is better; unit: PERCENT.

## Recorded results

| Method | mAP@0.5 (Gauss) | Result date | Report paper |
| --- | --- | --- | --- |
| Baron | 17.8 | Not recorded | 69cb555f0636e5bf4afca249 |
| PICA | 17.8 | Not recorded | 69cb555f0636e5bf4afca249 |
| Baron† | 17.7 | Not recorded | 69cb555f0636e5bf4afca249 |
| Obj | 16.1 | Not recorded | 69cb555f0636e5bf4afca249 |
| RALF | 16 | Not recorded | 69cb555f0636e5bf4afca249 |
| Baron\* | 15.8 | Not recorded | 69cb555f0636e5bf4afca249 |
| VL-PLM | 14.1 | Not recorded | 69cb555f0636e5bf4afca249 |
| R.CLIP | 12.5 | Not recorded | 69cb555f0636e5bf4afca249 |
| DQUO | 9.7 | Not recorded | 69cb555f0636e5bf4afca249 |
| SIA | 9.1 | Not recorded | 69cb555f0636e5bf4afca249 |
| CORA | 8.6 | Not recorded | 69cb555f0636e5bf4afca249 |

### Baron

mAP@0.5 (Gauss): 17.8. Result ID: 69cb998f47f34ce924b45701.

[Reporting paper](<https://www.sota2.com/research/paper/69cb555f0636e5bf4afca249.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa6eae1e209131cea41688.md>)

[Implementation](<https://github.com/wusize/ovdet>)

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

### PICA

mAP@0.5 (Gauss): 17.8. Result ID: 69cb998f47f34ce924b45707.

[Reporting paper](<https://www.sota2.com/research/paper/69cb555f0636e5bf4afca249.md>)

[Method origin paper](<https://www.sota2.com/research/paper/69cb555f0636e5bf4afca249.md>)

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

### Baron†

mAP@0.5 (Gauss): 17.7. Result ID: 69cb998f47f34ce924b45706.

[Reporting paper](<https://www.sota2.com/research/paper/69cb555f0636e5bf4afca249.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa62991e209131ce9fa0ca.md>)

Recorded evaluation configuration: {"Augmentation":"Mixup with DeepCORAL"}. This record does not assert independent reproduction.

### Obj

mAP@0.5 (Gauss): 16.1. Result ID: 69cb998f47f34ce924b456ff.

[Reporting paper](<https://www.sota2.com/research/paper/69cb555f0636e5bf4afca249.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa6e061e209131cea3d7e4.md>)

[Implementation](<https://github.com/hanoonaR/object-centric-ovd>)

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

### RALF

mAP@0.5 (Gauss): 16. Result ID: 69cb998f47f34ce924b45703.

[Reporting paper](<https://www.sota2.com/research/paper/69cb555f0636e5bf4afca249.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa72ce1e209131cea58981.md>)

[Implementation](<https://github.com/mlvlab/RALF>)

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

### Baron\*

mAP@0.5 (Gauss): 15.8. Result ID: 69cb998f47f34ce924b45705.

[Reporting paper](<https://www.sota2.com/research/paper/69cb555f0636e5bf4afca249.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa619e1e209131ce9f532d.md>)

Recorded evaluation configuration: {"Augmentation":"Mixup with DDC"}. This record does not assert independent reproduction.

### VL-PLM

mAP@0.5 (Gauss): 14.1. Result ID: 69cb998f47f34ce924b456fe.

[Reporting paper](<https://www.sota2.com/research/paper/69cb555f0636e5bf4afca249.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa6d171e209131cea37a18.md>)

[Implementation](<https://github.com/xiaofeng94/VL-PLM>)

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

### R.CLIP

mAP@0.5 (Gauss): 12.5. Result ID: 69cb998f47f34ce924b456fd.

[Reporting paper](<https://www.sota2.com/research/paper/69cb555f0636e5bf4afca249.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa6b7f1e209131cea2e983.md>)

[Implementation](<https://github.com/microsoft/RegionCLIP>)

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

### DQUO

mAP@0.5 (Gauss): 9.7. Result ID: 69cb998f47f34ce924b45704.

[Reporting paper](<https://www.sota2.com/research/paper/69cb555f0636e5bf4afca249.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa748d1e209131cea6267a.md>)

[Implementation](<https://github.com/xiaomoguhz/OV-DQUO>)

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

### SIA

mAP@0.5 (Gauss): 9.1. Result ID: 69cb998f47f34ce924b45702.

[Reporting paper](<https://www.sota2.com/research/paper/69cb555f0636e5bf4afca249.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa75221e209131cea660ae.md>)

[Implementation](<https://github.com/PKU-ICST-MIPL/SIA-OVD_ACMMM2024>)

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

### CORA

mAP@0.5 (Gauss): 8.6. Result ID: 69cb998f47f34ce924b45700.

[Reporting paper](<https://www.sota2.com/research/paper/69cb555f0636e5bf4afca249.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa6ecf1e209131cea424d1.md>)

[Implementation](<https://github.com/tgxs002/CORA>)

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

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