# Cuboid Autolabeling on KITTI Moderate (train)

Source: [Cuboid Autolabeling on KITTI Moderate (train)](<https://www.sota2.com/research/sota/cuboid-autolabeling-on-kitti-moderate-train>)

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

Data updated: 2026-02-26T05:36:43.441Z

Evaluation of 3D Object Autolabeling Quality performance specifically on the training split of the Moderate subset of the KITTI dataset, reporting BEV, 3D, and Novelty Score (NS) metrics.

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 Autolabeling Quality |
| Dataset (including any stated split) | KITTI Moderate (train) |
| Metric | BEV @ 0.5 |
| 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

- [BEV @ 0.5](<https://www.sota2.com/research/sota/cuboid-autolabeling-on-kitti-moderate-train.md?metric=BEV%20%40%200.5>): Higher is better; unit: PERCENT.

- [3D @ 0.5](<https://www.sota2.com/research/sota/cuboid-autolabeling-on-kitti-moderate-train.md?metric=3D%20%40%200.5>): Higher is better; unit: PERCENT.

- [NS @ 0.5](<https://www.sota2.com/research/sota/cuboid-autolabeling-on-kitti-moderate-train.md?metric=NS%20%40%200.5>): Higher is better; unit: PERCENT.

- [NS @ 1.0](<https://www.sota2.com/research/sota/cuboid-autolabeling-on-kitti-moderate-train.md?metric=NS%20%40%201.0>): Higher is better; unit: PERCENT.

## Recorded results

| Method | BEV @ 0.5 | Result date | Report paper |
| --- | --- | --- | --- |
| CSS-based Autolabeling | 63.36 | Not recorded | 68fa67561e209131cea159c9 |
| CSS-based Autolabeling | 61.17 | Not recorded | 68fa67561e209131cea159c9 |
| CSS-based Autolabeling | 59.75 | Not recorded | 68fa67561e209131cea159c9 |

### CSS-based Autolabeling

BEV @ 0.5: 63.36. Result ID: 698d2d938b517eb4331f935e.

[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: {"Input Source":"Mask-RCNN detections","Loop":"2"}. This record does not assert independent reproduction.

### CSS-based Autolabeling

BEV @ 0.5: 61.17. Result ID: 698d2d938b517eb4331f935d.

[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: {"Input Source":"RCNN detections","Loop":"2"}. This record does not assert independent reproduction.

### CSS-based Autolabeling

BEV @ 0.5: 59.75. Result ID: 698d2d938b517eb4331f935c.

[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: {"Input Source":"2D ground truth boxes","Loop":"2"}. This record does not assert independent reproduction.

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