# Depth Completion on Outdoor sequence KITTI → Waymo → VKITTI

Source: [Depth Completion on Outdoor sequence KITTI → Waymo → VKITTI](<https://www.sota2.com/research/sota/depth-completion-on-outdoor-sequence-kitti-waymo-vkitti>)

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

Data updated: 2026-02-27T10:11:29.464Z

Evaluation of Depth Completion performance on sequences derived from KITTI, Waymo, and VKITTI, utilizing specialized metrics focusing on Forgetting and SPTO performance.

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 | Depth Completion |
| Dataset (including any stated split) | Outdoor sequence KITTI → Waymo → VKITTI |
| Metric | Avg Forgetting MAE (%) |
| Unit | SCALAR |
| Direction | Lower is better |
| Results loaded | 21 of 21 |
| 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

- [Avg Forgetting MAE (%)](<https://www.sota2.com/research/sota/depth-completion-on-outdoor-sequence-kitti-waymo-vkitti.md?metric=Avg%20Forgetting%20MAE%20(%25)>): Lower is better; unit: SCALAR.

- [Avg Forgetting MAE](<https://www.sota2.com/research/sota/depth-completion-on-outdoor-sequence-kitti-waymo-vkitti.md?metric=Avg%20Forgetting%20MAE>): Lower is better; unit: SCALAR.

- [Avg Forgetting RMSE](<https://www.sota2.com/research/sota/depth-completion-on-outdoor-sequence-kitti-waymo-vkitti.md?metric=Avg%20Forgetting%20RMSE>): Lower is better; unit: SCALAR.

- [Avg Forgetting iMAE](<https://www.sota2.com/research/sota/depth-completion-on-outdoor-sequence-kitti-waymo-vkitti.md?metric=Avg%20Forgetting%20iMAE>): Lower is better; unit: SCALAR.

- [Avg Forgetting iRMSE](<https://www.sota2.com/research/sota/depth-completion-on-outdoor-sequence-kitti-waymo-vkitti.md?metric=Avg%20Forgetting%20iRMSE>): Lower is better; unit: SCALAR.

- [Avg Performance MAE](<https://www.sota2.com/research/sota/depth-completion-on-outdoor-sequence-kitti-waymo-vkitti.md?metric=Avg%20Performance%20MAE>): Lower is better; unit: SCALAR.

- [Avg Performance RMSE](<https://www.sota2.com/research/sota/depth-completion-on-outdoor-sequence-kitti-waymo-vkitti.md?metric=Avg%20Performance%20RMSE>): Lower is better; unit: SCALAR.

- [Avg Performance iMAE](<https://www.sota2.com/research/sota/depth-completion-on-outdoor-sequence-kitti-waymo-vkitti.md?metric=Avg%20Performance%20iMAE>): Lower is better; unit: SCALAR.

- [Avg Performance iRMSE](<https://www.sota2.com/research/sota/depth-completion-on-outdoor-sequence-kitti-waymo-vkitti.md?metric=Avg%20Performance%20iRMSE>): Lower is better; unit: SCALAR.

- [SPTO MAE](<https://www.sota2.com/research/sota/depth-completion-on-outdoor-sequence-kitti-waymo-vkitti.md?metric=SPTO%20MAE>): Lower is better; unit: SCALAR.

- [SPTO RMSE](<https://www.sota2.com/research/sota/depth-completion-on-outdoor-sequence-kitti-waymo-vkitti.md?metric=SPTO%20RMSE>): Lower is better; unit: SCALAR.

- [SPTO iMAE](<https://www.sota2.com/research/sota/depth-completion-on-outdoor-sequence-kitti-waymo-vkitti.md?metric=SPTO%20iMAE>): Lower is better; unit: SCALAR.

- [SPTO iRMSE](<https://www.sota2.com/research/sota/depth-completion-on-outdoor-sequence-kitti-waymo-vkitti.md?metric=SPTO%20iRMSE>): Lower is better; unit: SCALAR.

- [Avg Forgetting RMSE (%)](<https://www.sota2.com/research/sota/depth-completion-on-outdoor-sequence-kitti-waymo-vkitti.md?metric=Avg%20Forgetting%20RMSE%20(%25)>): Lower is better; unit: SCALAR.

- [Avg Forgetting iMAE (%)](<https://www.sota2.com/research/sota/depth-completion-on-outdoor-sequence-kitti-waymo-vkitti.md?metric=Avg%20Forgetting%20iMAE%20(%25)>): Lower is better; unit: SCALAR.

- [Avg Forgetting iRMSE (%)](<https://www.sota2.com/research/sota/depth-completion-on-outdoor-sequence-kitti-waymo-vkitti.md?metric=Avg%20Forgetting%20iRMSE%20(%25)>): Lower is better; unit: SCALAR.

- [Avg Performance MAE (mm)](<https://www.sota2.com/research/sota/depth-completion-on-outdoor-sequence-kitti-waymo-vkitti.md?metric=Avg%20Performance%20MAE%20(mm)>): Lower is better; unit: SCALAR.

- [Avg Performance RMSE (mm)](<https://www.sota2.com/research/sota/depth-completion-on-outdoor-sequence-kitti-waymo-vkitti.md?metric=Avg%20Performance%20RMSE%20(mm)>): Lower is better; unit: SCALAR.

- [Avg Performance iMAE (mm)](<https://www.sota2.com/research/sota/depth-completion-on-outdoor-sequence-kitti-waymo-vkitti.md?metric=Avg%20Performance%20iMAE%20(mm)>): Lower is better; unit: SCALAR.

- [Avg Performance iRMSE (mm)](<https://www.sota2.com/research/sota/depth-completion-on-outdoor-sequence-kitti-waymo-vkitti.md?metric=Avg%20Performance%20iRMSE%20(mm)>): Lower is better; unit: SCALAR.

- [SPTO MAE (mm)](<https://www.sota2.com/research/sota/depth-completion-on-outdoor-sequence-kitti-waymo-vkitti.md?metric=SPTO%20MAE%20(mm)>): Lower is better; unit: SCALAR.

- [SPTO RMSE (mm)](<https://www.sota2.com/research/sota/depth-completion-on-outdoor-sequence-kitti-waymo-vkitti.md?metric=SPTO%20RMSE%20(mm)>): Lower is better; unit: SCALAR.

- [SPTO iMAE (mm)](<https://www.sota2.com/research/sota/depth-completion-on-outdoor-sequence-kitti-waymo-vkitti.md?metric=SPTO%20iMAE%20(mm)>): Lower is better; unit: SCALAR.

- [SPTO iRMSE (mm)](<https://www.sota2.com/research/sota/depth-completion-on-outdoor-sequence-kitti-waymo-vkitti.md?metric=SPTO%20iRMSE%20(mm)>): Lower is better; unit: SCALAR.

## Recorded results

| Method | Avg Forgetting MAE (%) | Result date | Report paper |
| --- | --- | --- | --- |
| ProtoDepth | 0 | Not recorded | 68fa77741e209131cea7339d |
| ProtoDepth | 0 | Not recorded | 68fa77741e209131cea7339d |
| ProtoDepth | 0 | Not recorded | 68fa77741e209131cea7339d |
| ProtoDepth-A | 2.2 | Not recorded | 68fa77741e209131cea7339d |
| ProtoDepth-A | 2.427 | Not recorded | 68fa77741e209131cea7339d |
| ProtoDepth-A | 4.513 | Not recorded | 68fa77741e209131cea7339d |
| Replay | 8.29 | Not recorded | 68fa77741e209131cea7339d |
| Finetuned | 11.336 | Not recorded | 68fa77741e209131cea7339d |
| LwF | 12.368 | Not recorded | 68fa77741e209131cea7339d |
| Replay | 17.241 | Not recorded | 68fa77741e209131cea7339d |
| EWC | 21.006 | Not recorded | 68fa77741e209131cea7339d |
| LwF | 21.184 | Not recorded | 68fa77741e209131cea7339d |
| EWC | 23.517 | Not recorded | 68fa77741e209131cea7339d |
| Replay | 25.423 | Not recorded | 68fa77741e209131cea7339d |
| Finetuned | 27.153 | Not recorded | 68fa77741e209131cea7339d |
| Finetuned | 499.598 | Not recorded | 68fa77741e209131cea7339d |
| EWC | 555.925 | Not recorded | 68fa77741e209131cea7339d |
| LwF | 631.119 | Not recorded | 68fa77741e209131cea7339d |

### ProtoDepth

Avg Forgetting MAE (%): 0. Result ID: 69a16db835dfcabb00e6fa65.

[Reporting paper](<https://www.sota2.com/research/paper/68fa77741e209131cea7339d.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa77741e209131cea7339d.md>)

[Implementation](<https://protodepth.github.io/>)

Recorded evaluation configuration: {"Backbone":"VOICED","Strategy":"Incremental"}. This record does not assert independent reproduction.

### ProtoDepth

Avg Forgetting MAE (%): 0. Result ID: 69a16db835dfcabb00e6fa6b.

[Reporting paper](<https://www.sota2.com/research/paper/68fa77741e209131cea7339d.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa77741e209131cea7339d.md>)

[Implementation](<https://protodepth.github.io/>)

Recorded evaluation configuration: {"Backbone":"FusionNet","Strategy":"Incremental"}. This record does not assert independent reproduction.

### ProtoDepth

Avg Forgetting MAE (%): 0. Result ID: 69a16db835dfcabb00e6fa71.

[Reporting paper](<https://www.sota2.com/research/paper/68fa77741e209131cea7339d.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa77741e209131cea7339d.md>)

[Implementation](<https://protodepth.github.io/>)

Recorded evaluation configuration: {"Backbone":"KBNet","Strategy":"Incremental"}. This record does not assert independent reproduction.

### ProtoDepth-A

Avg Forgetting MAE (%): 2.2. Result ID: 69a16db835dfcabb00e6fa6a.

[Reporting paper](<https://www.sota2.com/research/paper/68fa77741e209131cea7339d.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa77741e209131cea7339d.md>)

[Implementation](<https://protodepth.github.io/>)

Recorded evaluation configuration: {"Backbone":"FusionNet","Strategy":"Agnostic"}. This record does not assert independent reproduction.

### ProtoDepth-A

Avg Forgetting MAE (%): 2.427. Result ID: 69a16db835dfcabb00e6fa64.

[Reporting paper](<https://www.sota2.com/research/paper/68fa77741e209131cea7339d.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa77741e209131cea7339d.md>)

[Implementation](<https://protodepth.github.io/>)

Recorded evaluation configuration: {"Backbone":"VOICED","Strategy":"Agnostic"}. This record does not assert independent reproduction.

### ProtoDepth-A

Avg Forgetting MAE (%): 4.513. Result ID: 69a16db835dfcabb00e6fa70.

[Reporting paper](<https://www.sota2.com/research/paper/68fa77741e209131cea7339d.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa77741e209131cea7339d.md>)

[Implementation](<https://protodepth.github.io/>)

Recorded evaluation configuration: {"Backbone":"KBNet","Strategy":"Agnostic"}. This record does not assert independent reproduction.

### Replay

Avg Forgetting MAE (%): 8.29. Result ID: 69a16db835dfcabb00e6fa69.

[Reporting paper](<https://www.sota2.com/research/paper/68fa77741e209131cea7339d.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa66b71e209131cea120af.md>)

Recorded evaluation configuration: {"Backbone":"FusionNet","Strategy":"Continual Learning"}. This record does not assert independent reproduction.

### Finetuned

Avg Forgetting MAE (%): 11.336. Result ID: 69a16db835dfcabb00e6fa66.

[Reporting paper](<https://www.sota2.com/research/paper/68fa77741e209131cea7339d.md>)

Recorded evaluation configuration: {"Backbone":"FusionNet","Strategy":"Continual Fine-tuning"}. This record does not assert independent reproduction.

### LwF

Avg Forgetting MAE (%): 12.368. Result ID: 69a16db835dfcabb00e6fa68.

[Reporting paper](<https://www.sota2.com/research/paper/68fa77741e209131cea7339d.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa63201e209131ce9fd0b3.md>)

Recorded evaluation configuration: {"Backbone":"FusionNet","Strategy":"Continual Learning"}. This record does not assert independent reproduction.

### Replay

Avg Forgetting MAE (%): 17.241. Result ID: 69a16db835dfcabb00e6fa63.

[Reporting paper](<https://www.sota2.com/research/paper/68fa77741e209131cea7339d.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa66b71e209131cea120af.md>)

Recorded evaluation configuration: {"Backbone":"VOICED","Strategy":"Continual Learning"}. This record does not assert independent reproduction.

### EWC

Avg Forgetting MAE (%): 21.006. Result ID: 69a16db835dfcabb00e6fa67.

[Reporting paper](<https://www.sota2.com/research/paper/68fa77741e209131cea7339d.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa6cb31e209131cea3555e.md>)

Recorded evaluation configuration: {"Backbone":"FusionNet","Strategy":"Continual Learning"}. This record does not assert independent reproduction.

### LwF

Avg Forgetting MAE (%): 21.184. Result ID: 69a16db835dfcabb00e6fa6e.

[Reporting paper](<https://www.sota2.com/research/paper/68fa77741e209131cea7339d.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa63201e209131ce9fd0b3.md>)

Recorded evaluation configuration: {"Backbone":"KBNet","Strategy":"Continual Learning"}. This record does not assert independent reproduction.

### EWC

Avg Forgetting MAE (%): 23.517. Result ID: 69a16db835dfcabb00e6fa6d.

[Reporting paper](<https://www.sota2.com/research/paper/68fa77741e209131cea7339d.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa6cb31e209131cea3555e.md>)

Recorded evaluation configuration: {"Backbone":"KBNet","Strategy":"Continual Learning"}. This record does not assert independent reproduction.

### Replay

Avg Forgetting MAE (%): 25.423. Result ID: 69a16db835dfcabb00e6fa6f.

[Reporting paper](<https://www.sota2.com/research/paper/68fa77741e209131cea7339d.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa66b71e209131cea120af.md>)

Recorded evaluation configuration: {"Backbone":"KBNet","Strategy":"Continual Learning"}. This record does not assert independent reproduction.

### Finetuned

Avg Forgetting MAE (%): 27.153. Result ID: 69a16db835dfcabb00e6fa6c.

[Reporting paper](<https://www.sota2.com/research/paper/68fa77741e209131cea7339d.md>)

Recorded evaluation configuration: {"Backbone":"KBNet","Strategy":"Continual Fine-tuning"}. This record does not assert independent reproduction.

### Finetuned

Avg Forgetting MAE (%): 499.598. Result ID: 69a16db835dfcabb00e6fa60.

[Reporting paper](<https://www.sota2.com/research/paper/68fa77741e209131cea7339d.md>)

Recorded evaluation configuration: {"Backbone":"VOICED","Strategy":"Continual Fine-tuning"}. This record does not assert independent reproduction.

### EWC

Avg Forgetting MAE (%): 555.925. Result ID: 69a16db835dfcabb00e6fa61.

[Reporting paper](<https://www.sota2.com/research/paper/68fa77741e209131cea7339d.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa6cb31e209131cea3555e.md>)

Recorded evaluation configuration: {"Backbone":"VOICED","Strategy":"Continual Learning"}. This record does not assert independent reproduction.

### LwF

Avg Forgetting MAE (%): 631.119. Result ID: 69a16db835dfcabb00e6fa62.

[Reporting paper](<https://www.sota2.com/research/paper/68fa77741e209131cea7339d.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa63201e209131ce9fd0b3.md>)

Recorded evaluation configuration: {"Backbone":"VOICED","Strategy":"Continual Learning"}. This record does not assert independent reproduction.

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