# Text-Conditioned 3D Shape Generation on ShapeNet and 3D Cartoon Monster mixture (test)

Source: [Text-Conditioned 3D Shape Generation on ShapeNet and 3D Cartoon Monster mixture (test)](<https://www.sota2.com/research/sota/text-conditioned-3d-shape-generation-on-shapenet-and-3d-cartoon-monster-mixture-test>)

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

Data updated: 2026-02-26T06:03:32.184Z

Evaluation of models generating 3D shapes conditioned on text prompts, using a mixed test set composed of ShapeNet models and 3D Cartoon Monsters. Evaluated using ST-S (ULIP), ST-S (SITA), P-FID, and P-IS metrics.

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 | Text-Conditioned 3D Shape Generation |
| Dataset (including any stated split) | ShapeNet and 3D Cartoon Monster mixture (test) |
| Metric | ST-S (ULIP) |
| Unit | SCALAR |
| Direction | Higher is better |
| Results loaded | 3 of 3 |
| 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

- [ST-S (ULIP)](<https://www.sota2.com/research/sota/text-conditioned-3d-shape-generation-on-shapenet-and-3d-cartoon-monster-mixture-test.md?metric=ST-S%20(ULIP)>): Higher is better; unit: SCALAR.

- [ST-S (SITA)](<https://www.sota2.com/research/sota/text-conditioned-3d-shape-generation-on-shapenet-and-3d-cartoon-monster-mixture-test.md?metric=ST-S%20(SITA)>): Higher is better; unit: SCALAR.

- [P-FID](<https://www.sota2.com/research/sota/text-conditioned-3d-shape-generation-on-shapenet-and-3d-cartoon-monster-mixture-test.md?metric=P-FID>): Lower is better; unit: SCALAR.

- [P-IS](<https://www.sota2.com/research/sota/text-conditioned-3d-shape-generation-on-shapenet-and-3d-cartoon-monster-mixture-test.md?metric=P-IS>): Higher is better; unit: SCALAR.

## Recorded results

| Method | ST-S (ULIP) | Result date | Report paper |
| --- | --- | --- | --- |
| Michelangelo | 16.647 | Not recorded | 68fa6fe61e209131cea47fc9 |
| 3DS2V | 12.934 | Not recorded | 68fa6fe61e209131cea47fc9 |
| 3DILG | 10.293 | Not recorded | 68fa6fe61e209131cea47fc9 |

### Michelangelo

ST-S (ULIP): 16.647. Result ID: 6990f0f13a6c314849047563.

[Reporting paper](<https://www.sota2.com/research/paper/68fa6fe61e209131cea47fc9.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa6fe61e209131cea47fc9.md>)

[Implementation](<https://github.com/NeuralCarver/Michelangelo>)

Recorded evaluation configuration: {"Conditioning":"Text-Conditioned"}. This record does not assert independent reproduction.

### 3DS2V

ST-S (ULIP): 12.934. Result ID: 6990f0f13a6c314849047562.

[Reporting paper](<https://www.sota2.com/research/paper/68fa6fe61e209131cea47fc9.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa6f371e209131cea4480d.md>)

Recorded evaluation configuration: {"Protocol":"Finetuned and Retrained","Conditioning":"Text-Conditioned"}. This record does not assert independent reproduction.

### 3DILG

ST-S (ULIP): 10.293. Result ID: 6990f0f13a6c314849047561.

[Reporting paper](<https://www.sota2.com/research/paper/68fa6fe61e209131cea47fc9.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa6da81e209131cea3b30e.md>)

Recorded evaluation configuration: {"Protocol":"Finetuned and Retrained","Conditioning":"Text-Conditioned"}. This record does not assert independent reproduction.

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