# Monocular 3D Object Detection on KITTI3D car AP40 (val)

Source: [Monocular 3D Object Detection on KITTI3D car AP40 (val)](<https://www.sota2.com/research/sota/monocular-3d-object-detection-on-kitti3d-car-ap40-val>)

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

Data updated: 2026-02-27T06:23:47.512Z

Evaluation of Monocular 3D Object Detection performance on the KITTI validation set, specifically focusing on the 'car' category using the AP40 protocol, reporting Average Precision (AP) for both 3D and Bird's Eye View (BEV) metrics across Easy, Moderate, and Hard difficulty levels, broken down by IoU thresholds of 0.5 and 0.7.

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 | Monocular 3D Object Detection |
| Dataset (including any stated split) | KITTI3D car AP40 (val) |
| Metric | AP (BEV) IoU=0.7 (Easy) |
| Unit | PERCENT |
| Direction | Higher is better |
| Results loaded | 7 of 7 |
| 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 (BEV) IoU=0.7 (Easy)](<https://www.sota2.com/research/sota/monocular-3d-object-detection-on-kitti3d-car-ap40-val.md?metric=AP%20(BEV)%20IoU%3D0.7%20(Easy)>): Higher is better; unit: PERCENT.

- [AP (BEV) IoU=0.5 (Easy)](<https://www.sota2.com/research/sota/monocular-3d-object-detection-on-kitti3d-car-ap40-val.md?metric=AP%20(BEV)%20IoU%3D0.5%20(Easy)>): Higher is better; unit: PERCENT.

- [AP (BEV) IoU=0.5 (Moderate)](<https://www.sota2.com/research/sota/monocular-3d-object-detection-on-kitti3d-car-ap40-val.md?metric=AP%20(BEV)%20IoU%3D0.5%20(Moderate)>): Higher is better; unit: PERCENT.

- [AP (BEV) IoU=0.5 (Hard)](<https://www.sota2.com/research/sota/monocular-3d-object-detection-on-kitti3d-car-ap40-val.md?metric=AP%20(BEV)%20IoU%3D0.5%20(Hard)>): Higher is better; unit: PERCENT.

- [AP (3D) IoU=0.5 (Easy)](<https://www.sota2.com/research/sota/monocular-3d-object-detection-on-kitti3d-car-ap40-val.md?metric=AP%20(3D)%20IoU%3D0.5%20(Easy)>): Higher is better; unit: PERCENT.

- [AP (3D) IoU=0.5 (Moderate)](<https://www.sota2.com/research/sota/monocular-3d-object-detection-on-kitti3d-car-ap40-val.md?metric=AP%20(3D)%20IoU%3D0.5%20(Moderate)>): Higher is better; unit: PERCENT.

- [AP (3D) IoU=0.5 (Hard)](<https://www.sota2.com/research/sota/monocular-3d-object-detection-on-kitti3d-car-ap40-val.md?metric=AP%20(3D)%20IoU%3D0.5%20(Hard)>): Higher is better; unit: PERCENT.

- [AP (BEV) IoU=0.7 (Moderate)](<https://www.sota2.com/research/sota/monocular-3d-object-detection-on-kitti3d-car-ap40-val.md?metric=AP%20(BEV)%20IoU%3D0.7%20(Moderate)>): Higher is better; unit: PERCENT.

- [AP (BEV) IoU=0.7 (Hard)](<https://www.sota2.com/research/sota/monocular-3d-object-detection-on-kitti3d-car-ap40-val.md?metric=AP%20(BEV)%20IoU%3D0.7%20(Hard)>): Higher is better; unit: PERCENT.

- [AP (3D) IoU=0.7 (Easy)](<https://www.sota2.com/research/sota/monocular-3d-object-detection-on-kitti3d-car-ap40-val.md?metric=AP%20(3D)%20IoU%3D0.7%20(Easy)>): Higher is better; unit: PERCENT.

- [AP (3D) IoU=0.7 (Moderate)](<https://www.sota2.com/research/sota/monocular-3d-object-detection-on-kitti3d-car-ap40-val.md?metric=AP%20(3D)%20IoU%3D0.7%20(Moderate)>): Higher is better; unit: PERCENT.

- [AP (3D) IoU=0.7 (Hard)](<https://www.sota2.com/research/sota/monocular-3d-object-detection-on-kitti3d-car-ap40-val.md?metric=AP%20(3D)%20IoU%3D0.7%20(Hard)>): Higher is better; unit: PERCENT.

## Recorded results

| Method | AP (BEV) IoU=0.7 (Easy) | Result date | Report paper |
| --- | --- | --- | --- |
| MonoPair | 24.12 | Not recorded | 68fa67221e209131cea14918 |
| +sigma\_z + sigma\_uv | 21.71 | Not recorded | 68fa67221e209131cea14918 |
| M3D-RPN | 20.85 | Not recorded | 68fa67221e209131cea14918 |
| Baseline | 19.83 | Not recorded | 68fa67221e209131cea14918 |
| MonoGRNet | 19.72 | Not recorded | 68fa67221e209131cea14918 |
| MonoDIS | 18.45 | Not recorded | 68fa67221e209131cea14918 |
| CenterNet | 3.46 | Not recorded | 68fa67221e209131cea14918 |

### MonoPair

AP (BEV) IoU=0.7 (Easy): 24.12. Result ID: 69a1385c35dfcabb00e596d4.

[Reporting paper](<https://www.sota2.com/research/paper/68fa67221e209131cea14918.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa67221e209131cea14918.md>)

Recorded evaluation configuration: {"RT (ms)":"57"}. This record does not assert independent reproduction.

### +sigma\_z + sigma\_uv

AP (BEV) IoU=0.7 (Easy): 21.71. Result ID: 69a1385c35dfcabb00e596d3.

[Reporting paper](<https://www.sota2.com/research/paper/68fa67221e209131cea14918.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa67221e209131cea14918.md>)

Recorded evaluation configuration: {"RT (ms)":"50"}. This record does not assert independent reproduction.

### M3D-RPN

AP (BEV) IoU=0.7 (Easy): 20.85. Result ID: 69a1385c35dfcabb00e596d1.

[Reporting paper](<https://www.sota2.com/research/paper/68fa67221e209131cea14918.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa663a1e209131cea0ef88.md>)

[Implementation](<http://cvlab.cse.msu.edu/project-m3d-rpn.html>)

Recorded evaluation configuration: {"RT (ms)":"161"}. This record does not assert independent reproduction.

### Baseline

AP (BEV) IoU=0.7 (Easy): 19.83. Result ID: 69a1385c35dfcabb00e596d2.

[Reporting paper](<https://www.sota2.com/research/paper/68fa67221e209131cea14918.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa67221e209131cea14918.md>)

Recorded evaluation configuration: {"RT (ms)":"47"}. This record does not assert independent reproduction.

### MonoGRNet

AP (BEV) IoU=0.7 (Easy): 19.72. Result ID: 69a1385c35dfcabb00e596d0.

[Reporting paper](<https://www.sota2.com/research/paper/68fa67221e209131cea14918.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa67581e209131cea15b29.md>)

[Implementation](<https://sites.google.com/view/monogrnet>)

Recorded evaluation configuration: {"RT (ms)":"60"}. This record does not assert independent reproduction.

### MonoDIS

AP (BEV) IoU=0.7 (Easy): 18.45. Result ID: 69a1385c35dfcabb00e596cf.

[Reporting paper](<https://www.sota2.com/research/paper/68fa67221e209131cea14918.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa65da1e209131cea0cd01.md>)

Recorded evaluation configuration: Not recorded. This record does not assert independent reproduction.

### CenterNet

AP (BEV) IoU=0.7 (Easy): 3.46. Result ID: 69a1385c35dfcabb00e596ce.

[Reporting paper](<https://www.sota2.com/research/paper/68fa67221e209131cea14918.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa65ab1e209131cea0bc9b.md>)

[Implementation](<https://github.com/xingyizhou/CenterNet>)

Recorded evaluation configuration: {"RT (ms)":"45"}. This record does not assert independent reproduction.

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