# 3D-consistent view synthesis on InfiniteNature-Zero evaluation set (5K images)

Source: [3D-consistent view synthesis on InfiniteNature-Zero evaluation set (5K images)](<https://www.sota2.com/research/sota/3d-consistent-view-synthesis-on-infinitenature-zero-evaluation-set-5k-images>)

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

Data updated: 2026-02-26T05:44:47.792Z

Evaluation of 3D-consistent view synthesis performance using metrics like FID (C\_forward), 1-step consistency, and cycle consistency on the InfiniteNature-Zero 5K image evaluation set.

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 | 3D-consistent view synthesis |
| Dataset (including any stated split) | InfiniteNature-Zero evaluation set (5K images) |
| Metric | FID (C\_forward) |
| Unit | SCALAR |
| Direction | Lower is better |
| Results loaded | 2 of 2 |
| 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

- [FID (C\_forward)](<https://www.sota2.com/research/sota/3d-consistent-view-synthesis-on-infinitenature-zero-evaluation-set-5k-images.md?metric=FID%20(C_forward)>): Lower is better; unit: SCALAR.

- [1-Step Consistency](<https://www.sota2.com/research/sota/3d-consistent-view-synthesis-on-infinitenature-zero-evaluation-set-5k-images.md?metric=1-Step%20Consistency>): Lower is better; unit: SCALAR.

- [Cycle Consistency](<https://www.sota2.com/research/sota/3d-consistent-view-synthesis-on-infinitenature-zero-evaluation-set-5k-images.md?metric=Cycle%20Consistency>): Lower is better; unit: SCALAR.

## Recorded results

| Method | FID (C\_forward) | Result date | Report paper |
| --- | --- | --- | --- |
| Persistent Nature | 26.09 | Not recorded | 68fa6edf1e209131cea428e5 |
| InfiniteNature-Zero | 28.15 | Not recorded | 68fa6edf1e209131cea428e5 |

### Persistent Nature

FID (C\_forward): 26.09. Result ID: 699065243a6c314849003e6c.

[Reporting paper](<https://www.sota2.com/research/paper/68fa6edf1e209131cea428e5.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa6edf1e209131cea428e5.md>)

[Implementation](<https://chail.github.io/persistent-nature/>)

Recorded evaluation configuration: {"Persistent":true,"Unbounded":true,"Resolution":"128px"}. This record does not assert independent reproduction.

### InfiniteNature-Zero

FID (C\_forward): 28.15. Result ID: 699065243a6c314849003e6b.

[Reporting paper](<https://www.sota2.com/research/paper/68fa6edf1e209131cea428e5.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa6d131e209131cea3776f.md>)

Recorded evaluation configuration: {"Persistent":false,"Unbounded":true}. This record does not assert independent reproduction.

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