# 3D Vehicle Detection on KITTI (val split 1)

Source: [3D Vehicle Detection on KITTI (val split 1)](<https://www.sota2.com/research/sota/3d-vehicle-detection-on-kitti-val-split-1>)

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

Data updated: 2026-02-26T05:23:07.455Z

Evaluation of 3D vehicle detection performance on the validation set split 1 of the KITTI dataset, reporting Average Precision (AP) 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 | Vehicle 3D Detection |
| Dataset (including any stated split) | KITTI (val split 1) |
| Metric | AP (Moderate) |
| Unit | PERCENT |
| Direction | Higher is better |
| Results loaded | 5 of 5 |
| 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 (Moderate)](<https://www.sota2.com/research/sota/3d-vehicle-detection-on-kitti-val-split-1.md?metric=AP%20(Moderate)>): Higher is better; unit: PERCENT.

- [AP (Easy)](<https://www.sota2.com/research/sota/3d-vehicle-detection-on-kitti-val-split-1.md?metric=AP%20(Easy)>): Higher is better; unit: PERCENT.

- [AP (Hard)](<https://www.sota2.com/research/sota/3d-vehicle-detection-on-kitti-val-split-1.md?metric=AP%20(Hard)>): Higher is better; unit: PERCENT.

## Recorded results

| Method | AP (Moderate) | Result date | Report paper |
| --- | --- | --- | --- |
| HR-NAS-B | 78.49 | Not recorded | 68fa6a291e209131cea2678e |
| Pointpillar | 77.12 | Not recorded | 68fa6a291e209131cea2678e |
| HR-NAS-A | 69.74 | Not recorded | 68fa6a291e209131cea2678e |
| MobileNetV2 | 67.65 | Not recorded | 68fa6a291e209131cea2678e |
| ShuffleNetV2 | 66.73 | Not recorded | 68fa6a291e209131cea2678e |

### HR-NAS-B

AP (Moderate): 78.49. Result ID: 698e1afdffc3901de09ad6a6.

[Reporting paper](<https://www.sota2.com/research/paper/68fa6a291e209131cea2678e.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa6a291e209131cea2678e.md>)

[Implementation](<https://github.com/dingmyu/HR-NAS>)

Recorded evaluation configuration: {"Params":"4.74M","FLOPs":"15.65G","Input size":"496 x 432","Framework":"Pointpillar \[43\]"}. This record does not assert independent reproduction.

### Pointpillar

AP (Moderate): 77.12. Result ID: 698e1afdffc3901de09ad6a4.

[Reporting paper](<https://www.sota2.com/research/paper/68fa6a291e209131cea2678e.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa65c01e209131cea0c2bc.md>)

[Implementation](<https://github.com/nutonomy/second.pytorch>)

Recorded evaluation configuration: {"Params":"4.80M","FLOPs":"61.75G","Input size":"496 x 432","Framework":"Pointpillar \[43\]"}. This record does not assert independent reproduction.

### HR-NAS-A

AP (Moderate): 69.74. Result ID: 698e1afdffc3901de09ad6a5.

[Reporting paper](<https://www.sota2.com/research/paper/68fa6a291e209131cea2678e.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa6a291e209131cea2678e.md>)

[Implementation](<https://github.com/dingmyu/HR-NAS>)

Recorded evaluation configuration: {"Params":"2.13M","FLOPs":"3.22G","Input size":"496 x 432","Framework":"Pointpillar \[43\]"}. This record does not assert independent reproduction.

### MobileNetV2

AP (Moderate): 67.65. Result ID: 698e1afdffc3901de09ad6a3.

[Reporting paper](<https://www.sota2.com/research/paper/68fa6a291e209131cea2678e.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa65811e209131cea0a9a1.md>)

[Implementation](<https://github.com/tensorflow/models/tree/master/research/slim/nets/mobilenet>)

Recorded evaluation configuration: {"Params":"2.49M","FLOPs":"6.47G","Input size":"496 x 432","Framework":"Pointpillar \[43\]"}. This record does not assert independent reproduction.

### ShuffleNetV2

AP (Moderate): 66.73. Result ID: 698e1afdffc3901de09ad6a2.

[Reporting paper](<https://www.sota2.com/research/paper/68fa6a291e209131cea2678e.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa64a61e209131cea05ddb.md>)

Recorded evaluation configuration: {"Params":"1.69M","FLOPs":"2.26G","Input size":"496 x 432","Framework":"Pointpillar \[43\]"}. This record does not assert independent reproduction.

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