# 3D Object Detection on Waymo Open Dataset (WOD) (val) (Class-Specific AP by L1/L2 and Mixed IoU)

Source: [3D Object Detection on Waymo Open Dataset (WOD) (val) (Class-Specific AP by L1/L2 and Mixed IoU)](<https://www.sota2.com/research/sota/3d-object-detection-on-waymo-open-dataset-wod-val-class-specific-ap-by-l1-l2-and-mixed-iou>)

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

Data updated: 2026-02-26T06:08:11.796Z

Evaluation of 3D Object Detection performance on the validation split of the Waymo Open Dataset (WOD). Metrics focus on Average Precision (AP) for Vehicle classes at IoU=0.7, and Pedestrian/Cyclist classes at IoU=0.5, reported across Level 1 (L1) and Level 2 (L2) difficulty settings, alongside overall mAP scores.

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) | Waymo Open Dataset (WOD) (val) |
| Metric | Vehicle 3D AP@0.7 L1 |
| Unit | SCALAR |
| 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

- [Vehicle 3D AP@0.7 L1](<https://www.sota2.com/research/sota/3d-object-detection-on-waymo-open-dataset-wod-val-class-specific-ap-by-l1-l2-and-mixed-iou.md?metric=Vehicle%203D%20AP%400.7%20L1>): Higher is better; unit: SCALAR.

- [Vehicle 3D AP@0.7 L2](<https://www.sota2.com/research/sota/3d-object-detection-on-waymo-open-dataset-wod-val-class-specific-ap-by-l1-l2-and-mixed-iou.md?metric=Vehicle%203D%20AP%400.7%20L2>): Higher is better; unit: SCALAR.

- [Pedestrian 3D AP@0.5 L1](<https://www.sota2.com/research/sota/3d-object-detection-on-waymo-open-dataset-wod-val-class-specific-ap-by-l1-l2-and-mixed-iou.md?metric=Pedestrian%203D%20AP%400.5%20L1>): Higher is better; unit: SCALAR.

- [Pedestrian 3D AP@0.5 L2](<https://www.sota2.com/research/sota/3d-object-detection-on-waymo-open-dataset-wod-val-class-specific-ap-by-l1-l2-and-mixed-iou.md?metric=Pedestrian%203D%20AP%400.5%20L2>): Higher is better; unit: SCALAR.

- [Cyclist 3D AP@0.5 L1](<https://www.sota2.com/research/sota/3d-object-detection-on-waymo-open-dataset-wod-val-class-specific-ap-by-l1-l2-and-mixed-iou.md?metric=Cyclist%203D%20AP%400.5%20L1>): Higher is better; unit: SCALAR.

- [Cyclist 3D AP@0.5 L2](<https://www.sota2.com/research/sota/3d-object-detection-on-waymo-open-dataset-wod-val-class-specific-ap-by-l1-l2-and-mixed-iou.md?metric=Cyclist%203D%20AP%400.5%20L2>): Higher is better; unit: SCALAR.

- [mAP L1](<https://www.sota2.com/research/sota/3d-object-detection-on-waymo-open-dataset-wod-val-class-specific-ap-by-l1-l2-and-mixed-iou.md?metric=mAP%20L1>): Higher is better; unit: SCALAR.

- [mAP L2](<https://www.sota2.com/research/sota/3d-object-detection-on-waymo-open-dataset-wod-val-class-specific-ap-by-l1-l2-and-mixed-iou.md?metric=mAP%20L2>): Higher is better; unit: SCALAR.

## Recorded results

| Method | Vehicle 3D AP@0.7 L1 | Result date | Report paper |
| --- | --- | --- | --- |
| OWL | 48.08 | Not recorded | 69365b6fd6b0dd3342934b2d |
| CPD | 37.4 | Not recorded | 69365b6fd6b0dd3342934b2d |
| OYSTER | 14.66 | Not recorded | 69365b6fd6b0dd3342934b2d |
| MODEST | 6.46 | Not recorded | 69365b6fd6b0dd3342934b2d |

### OWL

Vehicle 3D AP@0.7 L1: 48.08. Result ID: 699887b25698665ec4e91e9b.

[Reporting paper](<https://www.sota2.com/research/paper/69365b6fd6b0dd3342934b2d.md>)

[Method origin paper](<https://www.sota2.com/research/paper/69365b6fd6b0dd3342934b2d.md>)

Recorded evaluation configuration: {"Reference":"-","Learning Protocol":"Unsupervised"}. This record does not assert independent reproduction.

### CPD

Vehicle 3D AP@0.7 L1: 37.4. Result ID: 699887b25698665ec4e91e9a.

[Reporting paper](<https://www.sota2.com/research/paper/69365b6fd6b0dd3342934b2d.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa73e41e209131cea5eb63.md>)

[Implementation](<https://github.com/hailanyi/CPD>)

Recorded evaluation configuration: {"Reference":"CVPR2024","Learning Protocol":"Unsupervised"}. This record does not assert independent reproduction.

### OYSTER

Vehicle 3D AP@0.7 L1: 14.66. Result ID: 699887b25698665ec4e91e99.

[Reporting paper](<https://www.sota2.com/research/paper/69365b6fd6b0dd3342934b2d.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa710e1e209131cea4ef13.md>)

Recorded evaluation configuration: {"Reference":"CVPR2023","Learning Protocol":"Unsupervised"}. This record does not assert independent reproduction.

### MODEST

Vehicle 3D AP@0.7 L1: 6.46. Result ID: 699887b25698665ec4e91e98.

[Reporting paper](<https://www.sota2.com/research/paper/69365b6fd6b0dd3342934b2d.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa6c181e209131cea32213.md>)

[Implementation](<https://github.com/YurongYou/MODEST>)

Recorded evaluation configuration: {"Reference":"CVPR2022","Learning Protocol":"Unsupervised"}. This record does not assert independent reproduction.

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