# Affine-invariant depth estimation on Consolidated (KITTI, NYUv2, ETH3D, ScanNet, DIODE)

Source: [Affine-invariant depth estimation on Consolidated (KITTI, NYUv2, ETH3D, ScanNet, DIODE)](<https://www.sota2.com/research/sota/affine-invariant-depth-estimation-on-consolidated-kitti-nyuv2-eth3d-scannet-diode>)

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

Data updated: 2026-03-25T13:27:50.349Z

Evaluation of Affine-invariant Depth Estimation performance aggregated across the KITTI, NYUv2, ETH3D, ScanNet, and DIODE datasets.

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 | Affine-invariant depth estimation |
| Dataset (including any stated split) | Consolidated (KITTI, NYUv2, ETH3D, ScanNet, DIODE) |
| Metric | Average Ranking (All) |
| Unit | SCALAR |
| Direction | Lower is better |
| Results loaded | 27 of 27 |
| 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

- [Average Ranking (All)](<https://www.sota2.com/research/sota/affine-invariant-depth-estimation-on-consolidated-kitti-nyuv2-eth3d-scannet-diode.md?metric=Average%20Ranking%20(All)>): Lower is better; unit: SCALAR.

- [Average Ranking (Group)](<https://www.sota2.com/research/sota/affine-invariant-depth-estimation-on-consolidated-kitti-nyuv2-eth3d-scannet-diode.md?metric=Average%20Ranking%20(Group)>): Lower is better; unit: SCALAR.

- [Rank Score](<https://www.sota2.com/research/sota/affine-invariant-depth-estimation-on-consolidated-kitti-nyuv2-eth3d-scannet-diode.md?metric=Rank%20Score>): Lower is better; unit: SCALAR.

## Recorded results

| Method | Average Ranking (All) | Result date | Report paper |
| --- | --- | --- | --- |
| Iris (ours)\* | 3.1 | Not recorded | 69ba31850636e5bf4afc078c |
| DepthAnything | 4.3 | Not recorded | 69ba31850636e5bf4afc078c |
| Lotus-D\* | 4.4 | Not recorded | 69ba31850636e5bf4afc078c |
| DepthAnything V2 | 5.4 | Not recorded | 69ba31850636e5bf4afc078c |
| Diffusion-E2E-FT\* | 5.6 | Not recorded | 69ba31850636e5bf4afc078c |
| Lotus-G\* | 5.6 | Not recorded | 69ba31850636e5bf4afc078c |
| GenPercept\* | 6.1 | Not recorded | 69ba31850636e5bf4afc078c |
| Marigold\* | 7.4 | Not recorded | 69ba31850636e5bf4afc078c |
| Marigold(LCM)\* | 8.5 | Not recorded | 69ba31850636e5bf4afc078c |
| HDN | 9.6 | Not recorded | 69ba31850636e5bf4afc078c |
| DPT | 9.9 | Not recorded | 69ba31850636e5bf4afc078c |
| GeoWizard\* | 9.9 | Not recorded | 69ba31850636e5bf4afc078c |
| Omnidata | 12.2 | Not recorded | 69ba31850636e5bf4afc078c |
| LeRes | 12.3 | Not recorded | 69ba31850636e5bf4afc078c |
| MiDaS | 15.2 | Not recorded | 69ba31850636e5bf4afc078c |
| DiverseDepth | 15.6 | Not recorded | 69ba31850636e5bf4afc078c |

### Iris (ours)\*

Average Ranking (All): 3.1. Result ID: 69ba63cf47f34ce924a44f59.

[Reporting paper](<https://www.sota2.com/research/paper/69ba31850636e5bf4afc078c.md>)

[Method origin paper](<https://www.sota2.com/research/paper/69ba31850636e5bf4afc078c.md>)

Recorded evaluation configuration: {"Training Data":"59K + 100K","diffusion-based":"true"}. This record does not assert independent reproduction.

### DepthAnything

Average Ranking (All): 4.3. Result ID: 69ba63cf47f34ce924a44f50.

[Reporting paper](<https://www.sota2.com/research/paper/69ba31850636e5bf4afc078c.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa72cb1e209131cea58771.md>)

[Implementation](<https://depth-anything.github.io/>)

Recorded evaluation configuration: {"Training Data":"62.6M"}. This record does not assert independent reproduction.

### Lotus-D\*

Average Ranking (All): 4.4. Result ID: 69ba63cf47f34ce924a44f56.

[Reporting paper](<https://www.sota2.com/research/paper/69ba31850636e5bf4afc078c.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa76ad1e209131cea6e620.md>)

Recorded evaluation configuration: {"Training Data":"59K","diffusion-based":"true"}. This record does not assert independent reproduction.

### DepthAnything V2

Average Ranking (All): 5.4. Result ID: 69ba63cf47f34ce924a44f51.

[Reporting paper](<https://www.sota2.com/research/paper/69ba31850636e5bf4afc078c.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa75521e209131cea6728c.md>)

[Implementation](<https://depth-anything-v2.github.io/>)

Recorded evaluation configuration: {"Training Data":"62.6M"}. This record does not assert independent reproduction.

### Diffusion-E2E-FT\*

Average Ranking (All): 5.6. Result ID: 69ba63cf47f34ce924a44f52.

[Reporting paper](<https://www.sota2.com/research/paper/69ba31850636e5bf4afc078c.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa77871e209131cea73d9d.md>)

Recorded evaluation configuration: {"Training Data":"74K","diffusion-based":"true"}. This record does not assert independent reproduction.

### Lotus-G\*

Average Ranking (All): 5.6. Result ID: 69ba63cf47f34ce924a44f57.

[Reporting paper](<https://www.sota2.com/research/paper/69ba31850636e5bf4afc078c.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa76ad1e209131cea6e620.md>)

Recorded evaluation configuration: {"Training Data":"59K","diffusion-based":"true"}. This record does not assert independent reproduction.

### GenPercept\*

Average Ranking (All): 6.1. Result ID: 69ba63cf47f34ce924a44f58.

[Reporting paper](<https://www.sota2.com/research/paper/69ba31850636e5bf4afc078c.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa75fd1e209131cea6ac8b.md>)

Recorded evaluation configuration: {"Training Data":"90K","diffusion-based":"true"}. This record does not assert independent reproduction.

### Marigold\*

Average Ranking (All): 7.4. Result ID: 69ba63cf47f34ce924a44f55.

[Reporting paper](<https://www.sota2.com/research/paper/69ba31850636e5bf4afc078c.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa72ba1e209131cea58310.md>)

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

Recorded evaluation configuration: {"Training Data":"74K","diffusion-based":"true"}. This record does not assert independent reproduction.

### Marigold(LCM)\*

Average Ranking (All): 8.5. Result ID: 69ba63cf47f34ce924a44f54.

[Reporting paper](<https://www.sota2.com/research/paper/69ba31850636e5bf4afc078c.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa72ba1e209131cea58310.md>)

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

Recorded evaluation configuration: {"Training Data":"74K","diffusion-based":"true"}. This record does not assert independent reproduction.

### HDN

Average Ranking (All): 9.6. Result ID: 69ba63cf47f34ce924a44f4f.

[Reporting paper](<https://www.sota2.com/research/paper/69ba31850636e5bf4afc078c.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa6da91e209131cea3b3ac.md>)

Recorded evaluation configuration: {"Training Data":"300K"}. This record does not assert independent reproduction.

### DPT

Average Ranking (All): 9.9. Result ID: 69ba63cf47f34ce924a44f4e.

[Reporting paper](<https://www.sota2.com/research/paper/69ba31850636e5bf4afc078c.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa69a01e209131cea232d8.md>)

[Implementation](<https://github.com/intel-isl/DPT>)

Recorded evaluation configuration: {"Training Data":"1.4M"}. This record does not assert independent reproduction.

### GeoWizard\*

Average Ranking (All): 9.9. Result ID: 69ba63cf47f34ce924a44f53.

[Reporting paper](<https://www.sota2.com/research/paper/69ba31850636e5bf4afc078c.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa72981e209131cea57694.md>)

Recorded evaluation configuration: {"Training Data":"280K","diffusion-based":"true"}. This record does not assert independent reproduction.

### Omnidata

Average Ranking (All): 12.2. Result ID: 69ba63cf47f34ce924a44f4d.

[Reporting paper](<https://www.sota2.com/research/paper/69ba31850636e5bf4afc078c.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa6afe1e209131cea2b9b2.md>)

Recorded evaluation configuration: {"Training Data":"12.2M"}. This record does not assert independent reproduction.

### LeRes

Average Ranking (All): 12.3. Result ID: 69ba63cf47f34ce924a44f4c.

[Reporting paper](<https://www.sota2.com/research/paper/69ba31850636e5bf4afc078c.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa68ef1e209131cea1f121.md>)

[Implementation](<https://git.io/Depth>)

Recorded evaluation configuration: {"Training Data":"354K"}. This record does not assert independent reproduction.

### MiDaS

Average Ranking (All): 15.2. Result ID: 69ba63cf47f34ce924a44f4b.

[Reporting paper](<https://www.sota2.com/research/paper/69ba31850636e5bf4afc078c.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa68371e209131cea1adc8.md>)

Recorded evaluation configuration: {"Training Data":"2M"}. This record does not assert independent reproduction.

### DiverseDepth

Average Ranking (All): 15.6. Result ID: 69ba63cf47f34ce924a44f4a.

[Reporting paper](<https://www.sota2.com/research/paper/69ba31850636e5bf4afc078c.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa673c1e209131cea1516b.md>)

Recorded evaluation configuration: {"Training Data":"320K"}. This record does not assert independent reproduction.

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