# BEV Object Detection on KITTI Leaderboard official (test)

Source: [BEV Object Detection on KITTI Leaderboard official (test)](<https://www.sota2.com/research/sota/bev-object-detection-on-kitti-leaderboard-official-test>)

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

Data updated: 2026-02-26T05:57:45.121Z

Evaluation of Bird's Eye View (BEV) Object Detection performance on the official KITTI test set, reporting Average Precision (AP) specifically for Pedestrian and Cyclist classes across Easy, Moderate, and Hard difficulty levels.

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 | BEV Object Detection |
| Dataset (including any stated split) | KITTI Leaderboard official (test) |
| Metric | Pedestrian AP (Easy) |
| 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

- [Pedestrian AP (Easy)](<https://www.sota2.com/research/sota/bev-object-detection-on-kitti-leaderboard-official-test.md?metric=Pedestrian%20AP%20(Easy)>): Higher is better; unit: PERCENT.

- [Pedestrian AP (Moderate)](<https://www.sota2.com/research/sota/bev-object-detection-on-kitti-leaderboard-official-test.md?metric=Pedestrian%20AP%20(Moderate)>): Higher is better; unit: PERCENT.

- [Pedestrian AP (Hard)](<https://www.sota2.com/research/sota/bev-object-detection-on-kitti-leaderboard-official-test.md?metric=Pedestrian%20AP%20(Hard)>): Higher is better; unit: PERCENT.

- [Cyclist AP (Easy)](<https://www.sota2.com/research/sota/bev-object-detection-on-kitti-leaderboard-official-test.md?metric=Cyclist%20AP%20(Easy)>): Higher is better; unit: PERCENT.

- [Cyclist AP (Moderate)](<https://www.sota2.com/research/sota/bev-object-detection-on-kitti-leaderboard-official-test.md?metric=Cyclist%20AP%20(Moderate)>): Higher is better; unit: PERCENT.

- [Cyclist AP (Hard)](<https://www.sota2.com/research/sota/bev-object-detection-on-kitti-leaderboard-official-test.md?metric=Cyclist%20AP%20(Hard)>): Higher is better; unit: PERCENT.

## Recorded results

| Method | Pedestrian AP (Easy) | Result date | Report paper |
| --- | --- | --- | --- |
| DSGN | 26.61 | Not recorded | 68fa67521e209131cea15711 |
| M3D-RPN | 5.56 | Not recorded | 68fa67521e209131cea15711 |
| RT3DStereo | 4.72 | Not recorded | 68fa67521e209131cea15711 |

### DSGN

Pedestrian AP (Easy): 26.61. Result ID: 698d2ab78b517eb4331f7f37.

[Reporting paper](<https://www.sota2.com/research/paper/68fa67521e209131cea15711.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa67521e209131cea15711.md>)

[Implementation](<https://github.com/chenyilun95/DSGN>)

Recorded evaluation configuration: {"Modality":"Stereo"}. This record does not assert independent reproduction.

### M3D-RPN

Pedestrian AP (Easy): 5.56. Result ID: 698d2ab78b517eb4331f7f35.

[Reporting paper](<https://www.sota2.com/research/paper/68fa67521e209131cea15711.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa663a1e209131cea0ef88.md>)

[Implementation](<http://cvlab.cse.msu.edu/project-m3d-rpn.html>)

Recorded evaluation configuration: {"Modality":"Mono"}. This record does not assert independent reproduction.

### RT3DStereo

Pedestrian AP (Easy): 4.72. Result ID: 698d2ab78b517eb4331f7f36.

[Reporting paper](<https://www.sota2.com/research/paper/68fa67521e209131cea15711.md>)

Recorded evaluation configuration: {"Modality":"Stereo"}. This record does not assert independent reproduction.

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