# Cuboid Autolabeling on KITTI Easy

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

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

Data updated: 2026-02-26T05:36:43.806Z

Evaluation of Cuboid Autolabeling performance on the Easy subset of the KITTI dataset, reporting metrics related to Bird's Eye View (BEV) and 3D spatial accuracy.

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 | Cuboid Autolabeling |
| Dataset (including any stated split) | KITTI Easy |
| Metric | BEV Accuracy @ 0.5 |
| Unit | PERCENT |
| Direction | Higher is better |
| Results loaded | 18 of 18 |
| 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 Accuracy @ 0.5](<https://www.sota2.com/research/sota/cuboid-autolabeling-on-kitti-easy.md?metric=BEV%20Accuracy%20%40%200.5>): Higher is better; unit: PERCENT.

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

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

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

## Recorded results

| Method | BEV Accuracy @ 0.5 | Result date | Report paper |
| --- | --- | --- | --- |
| CSS Net | 80.7 | Not recorded | 68fa67561e209131cea159c9 |
| CSS-based Autolabeling | 80.7 | Not recorded | 68fa67561e209131cea159c9 |
| 3D Autolabeling Pipeline (CSS Net) | 80.7 | Not recorded | 68fa67561e209131cea159c9 |
| CSS Net | 80.57 | Not recorded | 68fa67561e209131cea159c9 |
| CSS-based Autolabeling | 80.57 | Not recorded | 68fa67561e209131cea159c9 |
| 3D Autolabeling Pipeline (CSS Net) | 80.57 | Not recorded | 68fa67561e209131cea159c9 |
| CSS Net | 78.46 | Not recorded | 68fa67561e209131cea159c9 |
| CSS-based Autolabeling | 78.46 | Not recorded | 68fa67561e209131cea159c9 |
| 3D Autolabeling Pipeline (CSS Net) | 78.46 | Not recorded | 68fa67561e209131cea159c9 |
| CSS Net | 78.45 | Not recorded | 68fa67561e209131cea159c9 |
| CSS-based Autolabeling | 78.45 | Not recorded | 68fa67561e209131cea159c9 |
| 3D Autolabeling Pipeline (CSS Net) | 78.45 | Not recorded | 68fa67561e209131cea159c9 |
| CSS Net | 78.09 | Not recorded | 68fa67561e209131cea159c9 |
| CSS-based Autolabeling | 78.09 | Not recorded | 68fa67561e209131cea159c9 |
| 3D Autolabeling Pipeline (CSS Net) | 78.09 | Not recorded | 68fa67561e209131cea159c9 |
| CSS Net | 77.84 | Not recorded | 68fa67561e209131cea159c9 |
| CSS-based Autolabeling | 77.84 | Not recorded | 68fa67561e209131cea159c9 |
| 3D Autolabeling Pipeline (CSS Net) | 77.84 | Not recorded | 68fa67561e209131cea159c9 |

### CSS Net

BEV Accuracy @ 0.5: 80.7. Result ID: 698d2d938b517eb4331f9348.

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

### CSS-based Autolabeling

BEV Accuracy @ 0.5: 80.7. Result ID: 698d2d938b517eb4331f935b.

[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.

### 3D Autolabeling Pipeline (CSS Net)

BEV Accuracy @ 0.5: 80.7. Result ID: 698d2d938b517eb4331f9364.

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

### CSS Net

BEV Accuracy @ 0.5: 80.57. Result ID: 698d2d938b517eb4331f9347.

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

### CSS-based Autolabeling

BEV Accuracy @ 0.5: 80.57. Result ID: 698d2d938b517eb4331f935a.

[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.

### 3D Autolabeling Pipeline (CSS Net)

BEV Accuracy @ 0.5: 80.57. Result ID: 698d2d938b517eb4331f9363.

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

### CSS Net

BEV Accuracy @ 0.5: 78.46. Result ID: 698d2d938b517eb4331f9345.

[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: {"Loop":"1","Input Source":"MASK-RCNN"}. This record does not assert independent reproduction.

### CSS-based Autolabeling

BEV Accuracy @ 0.5: 78.46. Result ID: 698d2d938b517eb4331f9358.

[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":"1"}. This record does not assert independent reproduction.

### 3D Autolabeling Pipeline (CSS Net)

BEV Accuracy @ 0.5: 78.46. Result ID: 698d2d938b517eb4331f9361.

[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: {"Loop":"1","Input Source":"Mask-RCNN"}. This record does not assert independent reproduction.

### CSS Net

BEV Accuracy @ 0.5: 78.45. Result ID: 698d2d938b517eb4331f9344.

[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: {"Loop":"1","Input Source":"RCNN"}. This record does not assert independent reproduction.

### CSS-based Autolabeling

BEV Accuracy @ 0.5: 78.45. Result ID: 698d2d938b517eb4331f9357.

[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":"1"}. This record does not assert independent reproduction.

### 3D Autolabeling Pipeline (CSS Net)

BEV Accuracy @ 0.5: 78.45. Result ID: 698d2d938b517eb4331f9360.

[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: {"Loop":"1","Input Source":"RCNN"}. This record does not assert independent reproduction.

### CSS Net

BEV Accuracy @ 0.5: 78.09. Result ID: 698d2d938b517eb4331f9343.

[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: {"Loop":"1","Input Source":"KITTI GT"}. This record does not assert independent reproduction.

### CSS-based Autolabeling

BEV Accuracy @ 0.5: 78.09. Result ID: 698d2d938b517eb4331f9356.

[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":"1"}. This record does not assert independent reproduction.

### 3D Autolabeling Pipeline (CSS Net)

BEV Accuracy @ 0.5: 78.09. Result ID: 698d2d938b517eb4331f935f.

[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: {"Loop":"1","Input Source":"KITTI GT"}. This record does not assert independent reproduction.

### CSS Net

BEV Accuracy @ 0.5: 77.84. Result ID: 698d2d938b517eb4331f9346.

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

### CSS-based Autolabeling

BEV Accuracy @ 0.5: 77.84. Result ID: 698d2d938b517eb4331f9359.

[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.

### 3D Autolabeling Pipeline (CSS Net)

BEV Accuracy @ 0.5: 77.84. Result ID: 698d2d938b517eb4331f9362.

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

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