# Object-Level Anomaly Segmentation on STU (val)

Source: [Object-Level Anomaly Segmentation on STU (val)](<https://www.sota2.com/research/sota/object-level-anomaly-segmentation-on-stu-val>)

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

Data updated: 2026-04-13T09:20:08.650Z

Evaluation of Object-Level Anomaly Segmentation performance on the validation split of the STU dataset.

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 | Object-Level Anomaly Segmentation |
| Dataset (including any stated split) | STU (val) |
| Metric | Recall (Q) |
| Unit | PERCENT |
| Direction | Higher is better |
| Results loaded | 8 of 8 |
| 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

- [Recall (Q)](<https://www.sota2.com/research/sota/object-level-anomaly-segmentation-on-stu-val.md?metric=Recall%20(Q)>): Higher is better; unit: PERCENT.

- [SQ](<https://www.sota2.com/research/sota/object-level-anomaly-segmentation-on-stu-val.md?metric=SQ>): Higher is better; unit: PERCENT.

- [RQ](<https://www.sota2.com/research/sota/object-level-anomaly-segmentation-on-stu-val.md?metric=RQ>): Higher is better; unit: PERCENT.

- [UQ](<https://www.sota2.com/research/sota/object-level-anomaly-segmentation-on-stu-val.md?metric=UQ>): Higher is better; unit: PERCENT.

- [PQ](<https://www.sota2.com/research/sota/object-level-anomaly-segmentation-on-stu-val.md?metric=PQ>): Higher is better; unit: PERCENT.

## Recorded results

| Method | Recall (Q) | Result date | Report paper |
| --- | --- | --- | --- |
| NDP-EE | 50.5 | Not recorded | 69dc78707d0172140ec789ef |
| NDP-Entropy | 33.6 | Not recorded | 69dc78707d0172140ec789ef |
| NDP-Energy | 26.8 | Not recorded | 69dc78707d0172140ec789ef |
| MaxLogit | 26.64 | Not recorded | 69dc78707d0172140ec789ef |
| RbA | 21.84 | Not recorded | 69dc78707d0172140ec789ef |
| Deep Ensemble | 17.7 | Not recorded | 69dc78707d0172140ec789ef |
| Void Classifier | 17.35 | Not recorded | 69dc78707d0172140ec789ef |
| MC Dropout | 3.54 | Not recorded | 69dc78707d0172140ec789ef |

### NDP-EE

Recall (Q): 50.5. Result ID: 69dcb46647f34ce924c3e62d.

[Reporting paper](<https://www.sota2.com/research/paper/69dc78707d0172140ec789ef.md>)

[Method origin paper](<https://www.sota2.com/research/paper/69dc78707d0172140ec789ef.md>)

Recorded evaluation configuration: {"Auxiliary OOD Data":"✓","Architecture":"Mask4Former"}. This record does not assert independent reproduction.

### NDP-Entropy

Recall (Q): 33.6. Result ID: 69dcb46647f34ce924c3e62c.

[Reporting paper](<https://www.sota2.com/research/paper/69dc78707d0172140ec789ef.md>)

[Method origin paper](<https://www.sota2.com/research/paper/69dc78707d0172140ec789ef.md>)

Recorded evaluation configuration: {"Auxiliary OOD Data":"✓","Architecture":"Mask4Former"}. This record does not assert independent reproduction.

### NDP-Energy

Recall (Q): 26.8. Result ID: 69dcb46647f34ce924c3e62b.

[Reporting paper](<https://www.sota2.com/research/paper/69dc78707d0172140ec789ef.md>)

[Method origin paper](<https://www.sota2.com/research/paper/69dc78707d0172140ec789ef.md>)

Recorded evaluation configuration: {"Auxiliary OOD Data":"✓","Architecture":"Mask4Former"}. This record does not assert independent reproduction.

### MaxLogit

Recall (Q): 26.64. Result ID: 69dcb46647f34ce924c3e628.

[Reporting paper](<https://www.sota2.com/research/paper/69dc78707d0172140ec789ef.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa6c811e209131cea34603.md>)

Recorded evaluation configuration: {"Auxiliary OOD Data":"✗","Architecture":"Mask4Former"}. This record does not assert independent reproduction.

### RbA

Recall (Q): 21.84. Result ID: 69dcb46647f34ce924c3e62a.

[Reporting paper](<https://www.sota2.com/research/paper/69dc78707d0172140ec789ef.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa6ee11e209131cea429d5.md>)

Recorded evaluation configuration: {"Auxiliary OOD Data":"✗","Architecture":"Mask4Former"}. This record does not assert independent reproduction.

### Deep Ensemble

Recall (Q): 17.7. Result ID: 69dcb46647f34ce924c3e626.

[Reporting paper](<https://www.sota2.com/research/paper/69dc78707d0172140ec789ef.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa63bc1e209131cea0084f.md>)

Recorded evaluation configuration: {"Auxiliary OOD Data":"✗","Architecture":"Mask4Former"}. This record does not assert independent reproduction.

### Void Classifier

Recall (Q): 17.35. Result ID: 69dcb46647f34ce924c3e629.

[Reporting paper](<https://www.sota2.com/research/paper/69dc78707d0172140ec789ef.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa6ad21e209131cea2a90a.md>)

[Implementation](<https://www.sota2.com/fishyscapes.com>)

Recorded evaluation configuration: {"Auxiliary OOD Data":"✓","Architecture":"Mask4Former"}. This record does not assert independent reproduction.

### MC Dropout

Recall (Q): 3.54. Result ID: 69dcb46647f34ce924c3e627.

[Reporting paper](<https://www.sota2.com/research/paper/69dc78707d0172140ec789ef.md>)

Recorded evaluation configuration: {"Auxiliary OOD Data":"✗","Architecture":"Mask4Former"}. This record does not assert independent reproduction.

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