# Open-Vocabulary Part Segmentation on Pascal-Part-116 (test)

Source: [Open-Vocabulary Part Segmentation on Pascal-Part-116 (test)](<https://www.sota2.com/research/sota/open-vocabulary-part-segmentation-on-pascal-part-116-test>)

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

Data updated: 2026-02-26T06:04:10.685Z

Evaluation of Open-Vocabulary Part Segmentation performance on the test split of the Pascal-Part-116 dataset, reporting Seen Recall, Unseen Recall, and Harmonic Recall.

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 | Open-Vocabulary Part Segmentation |
| Dataset (including any stated split) | Pascal-Part-116 (test) |
| Metric | Seen Recall |
| Unit | PERCENT |
| Direction | Higher is better |
| Results loaded | 5 of 5 |
| 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

- [Seen Recall](<https://www.sota2.com/research/sota/open-vocabulary-part-segmentation-on-pascal-part-116-test.md?metric=Seen%20Recall>): Higher is better; unit: PERCENT.

- [Unseen Recall](<https://www.sota2.com/research/sota/open-vocabulary-part-segmentation-on-pascal-part-116-test.md?metric=Unseen%20Recall>): Higher is better; unit: PERCENT.

- [Harmonic Recall](<https://www.sota2.com/research/sota/open-vocabulary-part-segmentation-on-pascal-part-116-test.md?metric=Harmonic%20Recall>): Higher is better; unit: PERCENT.

## Recorded results

| Method | Seen Recall | Result date | Report paper |
| --- | --- | --- | --- |
| ZSSeg+ | 65.47 | Not recorded | 68fa76101e209131cea6b388 |
| PartCLIPSeg | 58.97 | Not recorded | 68fa76101e209131cea6b388 |
| PartCLIPSeg | 58.46 | Not recorded | 68fa76101e209131cea6b388 |
| CAT-Seg | 56 | Not recorded | 68fa76101e209131cea6b388 |
| CLIPSeg | 55.71 | Not recorded | 68fa76101e209131cea6b388 |

### ZSSeg+

Seen Recall: 65.47. Result ID: 6994238f3a6c3148491b537b.

[Reporting paper](<https://www.sota2.com/research/paper/68fa76101e209131cea6b388.md>)

Recorded evaluation configuration: {"Protocol":"Oracle-Obj setting"}. This record does not assert independent reproduction.

### PartCLIPSeg

Seen Recall: 58.97. Result ID: 6994238f3a6c3148491b537e.

[Reporting paper](<https://www.sota2.com/research/paper/68fa76101e209131cea6b388.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa76101e209131cea6b388.md>)

[Implementation](<https://github.com/kaist-cvml-lab/part-clipseg>)

Recorded evaluation configuration: {"Protocol":"Oracle-Obj setting","losses":"without Lsep and Lenh"}. This record does not assert independent reproduction.

### PartCLIPSeg

Seen Recall: 58.46. Result ID: 6994238f3a6c3148491b537f.

[Reporting paper](<https://www.sota2.com/research/paper/68fa76101e209131cea6b388.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa76101e209131cea6b388.md>)

[Implementation](<https://github.com/kaist-cvml-lab/part-clipseg>)

Recorded evaluation configuration: {"Protocol":"Oracle-Obj setting","losses":"with Lsep and Lenh"}. This record does not assert independent reproduction.

### CAT-Seg

Seen Recall: 56. Result ID: 6994238f3a6c3148491b537d.

[Reporting paper](<https://www.sota2.com/research/paper/68fa76101e209131cea6b388.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa72b11e209131cea57ef4.md>)

[Implementation](<https://ku-cvlab.github.io/CAT-Seg/>)

Recorded evaluation configuration: {"Protocol":"Oracle-Obj setting"}. This record does not assert independent reproduction.

### CLIPSeg

Seen Recall: 55.71. Result ID: 6994238f3a6c3148491b537c.

[Reporting paper](<https://www.sota2.com/research/paper/68fa76101e209131cea6b388.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa6c2f1e209131cea32a66.md>)

[Implementation](<https://eckerlab.org/code/clipseg>)

Recorded evaluation configuration: {"Protocol":"Oracle-Obj setting"}. This record does not assert independent reproduction.

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