# Video Depth Estimation on Sintel 384x384 (test)

Source: [Video Depth Estimation on Sintel 384x384 (test)](<https://www.sota2.com/research/sota/video-depth-estimation-on-sintel-384x384-test>)

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

Data updated: 2026-04-10T00:01:13.817Z

Evaluation of video depth estimation performance on the 384x384 resolution test split of the Sintel dataset.

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 | Video Depth Estimation |
| Dataset (including any stated split) | Sintel 384x384 (test) |
| Metric | Abs Rel Error |
| Unit | SCALAR |
| Direction | Lower is better |
| Results loaded | 11 of 11 |
| 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

- [Abs Rel Error](<https://www.sota2.com/research/sota/video-depth-estimation-on-sintel-384x384-test.md?metric=Abs%20Rel%20Error>): Lower is better; unit: SCALAR.

- [Delta 1 Accuracy](<https://www.sota2.com/research/sota/video-depth-estimation-on-sintel-384x384-test.md?metric=Delta%201%20Accuracy>): Higher is better; unit: PERCENT.

- [TGSE (x100)](<https://www.sota2.com/research/sota/video-depth-estimation-on-sintel-384x384-test.md?metric=TGSE%20(x100)>): Lower is better; unit: SCALAR.

- [FPS](<https://www.sota2.com/research/sota/video-depth-estimation-on-sintel-384x384-test.md?metric=FPS>): Higher is better; unit: SCALAR.

- [Total Time (ms)](<https://www.sota2.com/research/sota/video-depth-estimation-on-sintel-384x384-test.md?metric=Total%20Time%20(ms)>): Lower is better; unit: TIME.

- [I2D Time (ms)](<https://www.sota2.com/research/sota/video-depth-estimation-on-sintel-384x384-test.md?metric=I2D%20Time%20(ms)>): Lower is better; unit: TIME.

- [D2V Time (ms)](<https://www.sota2.com/research/sota/video-depth-estimation-on-sintel-384x384-test.md?metric=D2V%20Time%20(ms)>): Lower is better; unit: TIME.

## Recorded results

| Method | Abs Rel Error | Result date | Report paper |
| --- | --- | --- | --- |
| VDPP | 0.312 | Not recorded | 69d732b17d0172140ec76c6a |
| VDPP | 0.319 | Not recorded | 69d732b17d0172140ec76c6a |
| VDA-L | 0.324 | Not recorded | 69d732b17d0172140ec76c6a |
| DAv2-L | 0.352 | Not recorded | 69d732b17d0172140ec76c6a |
| VDA-S | 0.355 | Not recorded | 69d732b17d0172140ec76c6a |
| DAv2-B | 0.358 | Not recorded | 69d732b17d0172140ec76c6a |
| DAv2-L + NVDS | 0.384 | Not recorded | 69d732b17d0172140ec76c6a |
| DAv2-B + NVDS | 0.387 | Not recorded | 69d732b17d0172140ec76c6a |
| DPT-L + NVDS | 0.471 | Not recorded | 69d732b17d0172140ec76c6a |
| DPT-L | 0.548 | Not recorded | 69d732b17d0172140ec76c6a |
| VDPP | 0.559 | Not recorded | 69d732b17d0172140ec76c6a |

### VDPP

Abs Rel Error: 0.312. Result ID: 69d83db547f34ce924c0302d.

[Reporting paper](<https://www.sota2.com/research/paper/69d732b17d0172140ec76c6a.md>)

[Method origin paper](<https://www.sota2.com/research/paper/69d732b17d0172140ec76c6a.md>)

Recorded evaluation configuration: {"Method Category":"Post-processing","Base Model/Backbone":"DAv2-B","Post-processing Module":"Ours"}. This record does not assert independent reproduction.

### VDPP

Abs Rel Error: 0.319. Result ID: 69d83db547f34ce924c0302e.

[Reporting paper](<https://www.sota2.com/research/paper/69d732b17d0172140ec76c6a.md>)

[Method origin paper](<https://www.sota2.com/research/paper/69d732b17d0172140ec76c6a.md>)

Recorded evaluation configuration: {"Method Category":"Post-processing","Base Model/Backbone":"DAv2-L","Post-processing Module":"Ours"}. This record does not assert independent reproduction.

### VDA-L

Abs Rel Error: 0.324. Result ID: 69d83db547f34ce924c03025.

[Reporting paper](<https://www.sota2.com/research/paper/69d732b17d0172140ec76c6a.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa79711e209131cea7da63.md>)

Recorded evaluation configuration: {"Method Category":"End-to-end"}. This record does not assert independent reproduction.

### DAv2-L

Abs Rel Error: 0.352. Result ID: 69d83db547f34ce924c03028.

[Reporting paper](<https://www.sota2.com/research/paper/69d732b17d0172140ec76c6a.md>)

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

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

Recorded evaluation configuration: {"Method Category":"Image to Depth"}. This record does not assert independent reproduction.

### VDA-S

Abs Rel Error: 0.355. Result ID: 69d83db547f34ce924c03024.

[Reporting paper](<https://www.sota2.com/research/paper/69d732b17d0172140ec76c6a.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa79711e209131cea7da63.md>)

Recorded evaluation configuration: {"Method Category":"End-to-end"}. This record does not assert independent reproduction.

### DAv2-B

Abs Rel Error: 0.358. Result ID: 69d83db547f34ce924c03027.

[Reporting paper](<https://www.sota2.com/research/paper/69d732b17d0172140ec76c6a.md>)

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

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

Recorded evaluation configuration: {"Method Category":"Image to Depth"}. This record does not assert independent reproduction.

### DAv2-L + NVDS

Abs Rel Error: 0.384. Result ID: 69d83db547f34ce924c0302b.

[Reporting paper](<https://www.sota2.com/research/paper/69d732b17d0172140ec76c6a.md>)

Recorded evaluation configuration: {"Method Category":"Post-processing","Base Model/Backbone":"DAv2-L","Post-processing Module":"NVDS"}. This record does not assert independent reproduction.

### DAv2-B + NVDS

Abs Rel Error: 0.387. Result ID: 69d83db547f34ce924c0302a.

[Reporting paper](<https://www.sota2.com/research/paper/69d732b17d0172140ec76c6a.md>)

Recorded evaluation configuration: {"Method Category":"Post-processing","Base Model/Backbone":"DAv2-B","Post-processing Module":"NVDS"}. This record does not assert independent reproduction.

### DPT-L + NVDS

Abs Rel Error: 0.471. Result ID: 69d83db547f34ce924c03029.

[Reporting paper](<https://www.sota2.com/research/paper/69d732b17d0172140ec76c6a.md>)

Recorded evaluation configuration: {"Method Category":"Post-processing","Base Model/Backbone":"DPT-L","Post-processing Module":"NVDS"}. This record does not assert independent reproduction.

### DPT-L

Abs Rel Error: 0.548. Result ID: 69d83db547f34ce924c03026.

[Reporting paper](<https://www.sota2.com/research/paper/69d732b17d0172140ec76c6a.md>)

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

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

Recorded evaluation configuration: {"Method Category":"Image to Depth"}. This record does not assert independent reproduction.

### VDPP

Abs Rel Error: 0.559. Result ID: 69d83db547f34ce924c0302c.

[Reporting paper](<https://www.sota2.com/research/paper/69d732b17d0172140ec76c6a.md>)

[Method origin paper](<https://www.sota2.com/research/paper/69d732b17d0172140ec76c6a.md>)

Recorded evaluation configuration: {"Method Category":"Post-processing","Base Model/Backbone":"DPT-L","Post-processing Module":"Ours"}. This record does not assert independent reproduction.

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