# Object Detection on LVISdense

Source: [Object Detection on LVISdense](<https://www.sota2.com/research/sota/object-detection-on-lvisdense>)

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

Data updated: 2026-06-09T11:04:28.532Z

Evaluation of Object Detection performance on the LVISdense dataset, reporting metrics related to object size categories (Small, Medium, Large).

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 | Object Detection |
| Dataset (including any stated split) | LVISdense |
| Metric | AP (Small) |
| 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

- [AP (Small)](<https://www.sota2.com/research/sota/object-detection-on-lvisdense.md?metric=AP%20(Small)>): Higher is better; unit: PERCENT.

- [AP (Medium)](<https://www.sota2.com/research/sota/object-detection-on-lvisdense.md?metric=AP%20(Medium)>): Higher is better; unit: PERCENT.

- [AP (Large)](<https://www.sota2.com/research/sota/object-detection-on-lvisdense.md?metric=AP%20(Large)>): Higher is better; unit: PERCENT.

## Recorded results

| Method | AP (Small) | Result date | Report paper |
| --- | --- | --- | --- |
| GPT-5 | 11.3 | Not recorded | 68fa78d11e209131cea7a934 |
| ACTIVE-o3 | 7 | Not recorded | 68fa78d11e209131cea7a934 |
| CRG | 4.3 | Not recorded | 68fa78d11e209131cea7a934 |
| GPT-5 | 0.1 | Not recorded | 68fa78d11e209131cea7a934 |

### GPT-5

AP (Small): 11.3. Result ID: 6a27f1fbfbc373592c259a33.

[Reporting paper](<https://www.sota2.com/research/paper/68fa78d11e209131cea7a934.md>)

Recorded evaluation configuration: {"Grounding module":"GDINO"}. This record does not assert independent reproduction.

### ACTIVE-o3

AP (Small): 7. Result ID: 6a27f1fbfbc373592c259a31.

[Reporting paper](<https://www.sota2.com/research/paper/68fa78d11e209131cea7a934.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa78d11e209131cea7a934.md>)

Recorded evaluation configuration: {"Grounding module":"GDINO"}. This record does not assert independent reproduction.

### CRG

AP (Small): 4.3. Result ID: 6a27f1fbfbc373592c259a30.

[Reporting paper](<https://www.sota2.com/research/paper/68fa78d11e209131cea7a934.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa72681e209131cea564c7.md>)

Recorded evaluation configuration: {"Grounding module":"GDINO"}. This record does not assert independent reproduction.

### GPT-5

AP (Small): 0.1. Result ID: 6a27f1fbfbc373592c259a32.

[Reporting paper](<https://www.sota2.com/research/paper/68fa78d11e209131cea7a934.md>)

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.
