# 3D Object Detection on PandaSet (held-out)

Source: [3D Object Detection on PandaSet (held-out)](<https://www.sota2.com/research/sota/3d-object-detection-on-pandaset-held-out>)

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

Data updated: 2026-05-22T08:56:35.299Z

Evaluation of 3D Object Detection performance on the held-out split of the PandaSet dataset, reporting metrics including mAP and distance-specific APs (AP@1m, AP@2m, AP@4m).

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 | 3D Object Detection |
| Dataset (including any stated split) | PandaSet (held-out) |
| Metric | mAP |
| Unit | PERCENT |
| Direction | Higher is better |
| Results loaded | 3 of 3 |
| 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](<https://www.sota2.com/research/sota/3d-object-detection-on-pandaset-held-out.md?metric=mAP>): Higher is better; unit: PERCENT.

- [AP@1m](<https://www.sota2.com/research/sota/3d-object-detection-on-pandaset-held-out.md?metric=AP%401m>): Higher is better; unit: PERCENT.

- [AP@2m](<https://www.sota2.com/research/sota/3d-object-detection-on-pandaset-held-out.md?metric=AP%402m>): Higher is better; unit: PERCENT.

- [AP@4m](<https://www.sota2.com/research/sota/3d-object-detection-on-pandaset-held-out.md?metric=AP%404m>): Higher is better; unit: PERCENT.

## Recorded results

| Method | mAP | Result date | Report paper |
| --- | --- | --- | --- |
| Real + Sim (ours) | 27.7 | Not recorded | 6a0fe30af984739ed6739d63 |
| Real + Sim (3DGS) | 25.8 | Not recorded | 6a0fe30af984739ed6739d63 |
| Real | 25.6 | Not recorded | 6a0fe30af984739ed6739d63 |

### Real + Sim (ours)

mAP: 27.7. Result ID: 6a101a2bfbc373592c093e72.

[Reporting paper](<https://www.sota2.com/research/paper/6a0fe30af984739ed6739d63.md>)

[Method origin paper](<https://www.sota2.com/research/paper/6a0fe30af984739ed6739d63.md>)

Recorded evaluation configuration: {"Training Strategy":"Data Augmentation","Simulation Backbone":"GenRe+"}. This record does not assert independent reproduction.

### Real + Sim (3DGS)

mAP: 25.8. Result ID: 6a101a2bfbc373592c093e71.

[Reporting paper](<https://www.sota2.com/research/paper/6a0fe30af984739ed6739d63.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa70461e209131cea4a191.md>)

[Implementation](<https://repo-sam.inria.fr/fungraph/3d-gaussian-splatting/>)

Recorded evaluation configuration: {"Training Strategy":"Data Augmentation","Simulation Backbone":"3DGS"}. This record does not assert independent reproduction.

### Real

mAP: 25.6. Result ID: 6a101a2bfbc373592c093e70.

[Reporting paper](<https://www.sota2.com/research/paper/6a0fe30af984739ed6739d63.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa6b791e209131cea2e664.md>)

Recorded evaluation configuration: {"Training Strategy":"Real only"}. This record does not assert independent reproduction.

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