# Bird's Eye View (BEV) Object Detection on KITTI MV3D (val)

Source: [Bird's Eye View (BEV) Object Detection on KITTI MV3D (val)](<https://www.sota2.com/research/sota/bird-s-eye-view-bev-object-detection-on-kitti-mv3d-val>)

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

Data updated: 2026-02-26T05:36:43.757Z

Evaluation of Bird's Eye View (BEV) Object Detection performance on the validation split of the KITTI MV3D dataset, reporting Average Precision (AP) across IoU thresholds of 0.5 and 0.7, specifically for Easy and Moderate 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 | Bird's Eye View (BEV) Object Detection |
| Dataset (including any stated split) | KITTI MV3D (val) |
| Metric | BEV AP @ 0.5 (Easy) |
| 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

- [BEV AP @ 0.5 (Easy)](<https://www.sota2.com/research/sota/bird-s-eye-view-bev-object-detection-on-kitti-mv3d-val.md?metric=BEV%20AP%20%40%200.5%20(Easy)>): Higher is better; unit: PERCENT.

- [BEV AP @ 0.7 (Easy)](<https://www.sota2.com/research/sota/bird-s-eye-view-bev-object-detection-on-kitti-mv3d-val.md?metric=BEV%20AP%20%40%200.7%20(Easy)>): Higher is better; unit: PERCENT.

- [BEV AP @ 0.5 (Moderate)](<https://www.sota2.com/research/sota/bird-s-eye-view-bev-object-detection-on-kitti-mv3d-val.md?metric=BEV%20AP%20%40%200.5%20(Moderate)>): Higher is better; unit: PERCENT.

- [BEV AP @ 0.7 (Moderate)](<https://www.sota2.com/research/sota/bird-s-eye-view-bev-object-detection-on-kitti-mv3d-val.md?metric=BEV%20AP%20%40%200.7%20(Moderate)>): Higher is better; unit: PERCENT.

## Recorded results

| Method | BEV AP @ 0.5 (Easy) | Result date | Report paper |
| --- | --- | --- | --- |
| PointPillars | 95.1 | Not recorded | 68fa67561e209131cea159c9 |
| PointPillars | 94.9 | Not recorded | 68fa67561e209131cea159c9 |
| MonoDIS | 52.4 | Not recorded | 68fa67561e209131cea159c9 |
| MonoDIS | 51.1 | Not recorded | 68fa67561e209131cea159c9 |

### PointPillars

BEV AP @ 0.5 (Easy): 95.1. Result ID: 698d2d938b517eb4331f9352.

[Reporting paper](<https://www.sota2.com/research/paper/68fa67561e209131cea159c9.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa67561e209131cea159c9.md>)

[Implementation](<http://github.com/TRI-ML/sdflabel>)

Recorded evaluation configuration: {"Training Label Source":"Original Labels"}. This record does not assert independent reproduction.

### PointPillars

BEV AP @ 0.5 (Easy): 94.9. Result ID: 698d2d938b517eb4331f9353.

[Reporting paper](<https://www.sota2.com/research/paper/68fa67561e209131cea159c9.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa67561e209131cea159c9.md>)

[Implementation](<http://github.com/TRI-ML/sdflabel>)

Recorded evaluation configuration: {"Training Label Source":"Autolabels"}. This record does not assert independent reproduction.

### MonoDIS

BEV AP @ 0.5 (Easy): 52.4. Result ID: 698d2d938b517eb4331f9354.

[Reporting paper](<https://www.sota2.com/research/paper/68fa67561e209131cea159c9.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa67561e209131cea159c9.md>)

[Implementation](<http://github.com/TRI-ML/sdflabel>)

Recorded evaluation configuration: {"Training Label Source":"Original Labels"}. This record does not assert independent reproduction.

### MonoDIS

BEV AP @ 0.5 (Easy): 51.1. Result ID: 698d2d938b517eb4331f9355.

[Reporting paper](<https://www.sota2.com/research/paper/68fa67561e209131cea159c9.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa67561e209131cea159c9.md>)

[Implementation](<http://github.com/TRI-ML/sdflabel>)

Recorded evaluation configuration: {"Training Label Source":"Autolabels"}. This record does not assert independent reproduction.

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