# 3D Reconstruction on OmniObject3D (1700 unseen samples)

Source: [3D Reconstruction on OmniObject3D (1700 unseen samples)](<https://www.sota2.com/research/sota/3d-reconstruction-on-omniobject3d-1700-unseen-samples>)

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

Data updated: 2026-02-26T05:38:31.472Z

Evaluation of 3D Reconstruction performance on the specific 1700 unseen samples subset of the OmniObject3D dataset, reporting metrics including CLIP score, PSNR, and geometric fidelity measures (CD, FS@0.05, FS@0.1).

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 Reconstruction |
| Dataset (including any stated split) | OmniObject3D (1700 unseen samples) |
| Metric | CLIP Score |
| Unit | SCALAR |
| 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

- [CLIP Score](<https://www.sota2.com/research/sota/3d-reconstruction-on-omniobject3d-1700-unseen-samples.md?metric=CLIP%20Score>): Higher is better; unit: SCALAR.

- [PSNR](<https://www.sota2.com/research/sota/3d-reconstruction-on-omniobject3d-1700-unseen-samples.md?metric=PSNR>): Higher is better; unit: SCALAR.

- [CD](<https://www.sota2.com/research/sota/3d-reconstruction-on-omniobject3d-1700-unseen-samples.md?metric=CD>): Lower is better; unit: SCALAR.

- [FS@0.05](<https://www.sota2.com/research/sota/3d-reconstruction-on-omniobject3d-1700-unseen-samples.md?metric=FS%400.05>): Higher is better; unit: PERCENT.

- [FS@0.1](<https://www.sota2.com/research/sota/3d-reconstruction-on-omniobject3d-1700-unseen-samples.md?metric=FS%400.1>): Higher is better; unit: PERCENT.

## Recorded results

| Method | CLIP Score | Result date | Report paper |
| --- | --- | --- | --- |
| TriMM | 0.674 | Not recorded | 68fa7ad41e209131cea855d9 |
| Trellis | 0.673 | Not recorded | 68fa7ad41e209131cea855d9 |
| CRM | 0.672 | Not recorded | 68fa7ad41e209131cea855d9 |
| TripoSR | 0.664 | Not recorded | 68fa7ad41e209131cea855d9 |
| LGM | 0.663 | Not recorded | 68fa7ad41e209131cea855d9 |
| InstantMesh | 0.661 | Not recorded | 68fa7ad41e209131cea855d9 |
| Baseline | 0.653 | Not recorded | 68fa7ad41e209131cea855d9 |

### TriMM

CLIP Score: 0.674. Result ID: 699e1bb35698665ec4d4db0e.

[Reporting paper](<https://www.sota2.com/research/paper/68fa7ad41e209131cea855d9.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa7ad41e209131cea855d9.md>)

Recorded evaluation configuration: {"Training samples":"80K"}. This record does not assert independent reproduction.

### Trellis

CLIP Score: 0.673. Result ID: 699e1bb35698665ec4d4db0d.

[Reporting paper](<https://www.sota2.com/research/paper/68fa7ad41e209131cea855d9.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa78f21e209131cea7b608.md>)

[Implementation](<https://trellis3d.github.io/>)

Recorded evaluation configuration: {"Training samples":"500K"}. This record does not assert independent reproduction.

### CRM

CLIP Score: 0.672. Result ID: 699e1bb35698665ec4d4db0a.

[Reporting paper](<https://www.sota2.com/research/paper/68fa7ad41e209131cea855d9.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa727b1e209131cea56ad8.md>)

Recorded evaluation configuration: {"Training samples":"376K"}. This record does not assert independent reproduction.

### TripoSR

CLIP Score: 0.664. Result ID: 699e1bb35698665ec4d4db09.

[Reporting paper](<https://www.sota2.com/research/paper/68fa7ad41e209131cea855d9.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa72681e209131cea564a8.md>)

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

### LGM

CLIP Score: 0.663. Result ID: 699e1bb35698665ec4d4db0b.

[Reporting paper](<https://www.sota2.com/research/paper/68fa7ad41e209131cea855d9.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa72291e209131cea54dd3.md>)

Recorded evaluation configuration: {"Training samples":"80K"}. This record does not assert independent reproduction.

### InstantMesh

CLIP Score: 0.661. Result ID: 699e1bb35698665ec4d4db0c.

[Reporting paper](<https://www.sota2.com/research/paper/68fa7ad41e209131cea855d9.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa72da1e209131cea58e96.md>)

Recorded evaluation configuration: {"Training samples":"270K"}. This record does not assert independent reproduction.

### Baseline

CLIP Score: 0.653. Result ID: 699e1bb35698665ec4d4db08.

[Reporting paper](<https://www.sota2.com/research/paper/68fa7ad41e209131cea855d9.md>)

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

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