# Part-level recognition on PartImageNet

Source: [Part-level recognition on PartImageNet](<https://www.sota2.com/research/sota/part-level-recognition-on-partimagenet>)

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

Data updated: 2026-02-26T05:39:46.893Z

Evaluation of Part-level recognition performance on the PartImageNet dataset, reporting metrics such as Con., Sta., and Acc.

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 | Part-level recognition |
| Dataset (including any stated split) | PartImageNet |
| Metric | Consistency |
| Unit | PERCENT |
| Direction | Higher is better |
| Results loaded | 4 of 4 |
| 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

- [Consistency](<https://www.sota2.com/research/sota/part-level-recognition-on-partimagenet.md?metric=Consistency>): Higher is better; unit: PERCENT.

- [Stability](<https://www.sota2.com/research/sota/part-level-recognition-on-partimagenet.md?metric=Stability>): Higher is better; unit: PERCENT.

- [Accuracy](<https://www.sota2.com/research/sota/part-level-recognition-on-partimagenet.md?metric=Accuracy>): Higher is better; unit: PERCENT.

## Recorded results

| Method | Consistency | Result date | Report paper |
| --- | --- | --- | --- |
| IVPT | 63.2 | Not recorded | 68fa775f1e209131cea72cfd |
| Huang et al. | 58.6 | Not recorded | 68fa775f1e209131cea72cfd |
| VPT-Deep | 54.2 | Not recorded | 68fa775f1e209131cea72cfd |
| ProtoPool | 47.5 | Not recorded | 68fa775f1e209131cea72cfd |

### IVPT

Consistency: 63.2. Result ID: 699e14645698665ec4ccbcfc.

[Reporting paper](<https://www.sota2.com/research/paper/68fa775f1e209131cea72cfd.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa775f1e209131cea72cfd.md>)

Recorded evaluation configuration: {"Backbone":"DinoV2-B"}. This record does not assert independent reproduction.

### Huang et al.

Consistency: 58.6. Result ID: 699e14645698665ec4ccbcfa.

[Reporting paper](<https://www.sota2.com/research/paper/68fa775f1e209131cea72cfd.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa71071e209131cea4eb73.md>)

Recorded evaluation configuration: {"Backbone":"DinoV2-B"}. This record does not assert independent reproduction.

### VPT-Deep

Consistency: 54.2. Result ID: 699e14645698665ec4ccbcfb.

[Reporting paper](<https://www.sota2.com/research/paper/68fa775f1e209131cea72cfd.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa6d181e209131cea37ae9.md>)

[Implementation](<https://www.sota2.com/github.com/kmnp/vpt>)

Recorded evaluation configuration: {"Backbone":"DinoV2-B"}. This record does not assert independent reproduction.

### ProtoPool

Consistency: 47.5. Result ID: 699e14645698665ec4ccbcf9.

[Reporting paper](<https://www.sota2.com/research/paper/68fa775f1e209131cea72cfd.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa6d5a1e209131cea39665.md>)

Recorded evaluation configuration: {"Backbone":"DinoV2-B"}. This record does not assert independent reproduction.

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