# Building Change Detection on LEVIR building change detection (test)

Source: [Building Change Detection on LEVIR building change detection (test)](<https://www.sota2.com/research/sota/building-change-detection-on-levir-building-change-detection-test>)

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

Data updated: 2026-02-26T05:58:43.223Z

Evaluation of Building Change Detection performance on the test split of the LEVIR building change detection dataset, reporting Precision, Recall, F1, MCC, and IoU.

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 | Building Change Detection |
| Dataset (including any stated split) | LEVIR building change detection (test) |
| Metric | Precision |
| Unit | PERCENT |
| 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

- [Precision](<https://www.sota2.com/research/sota/building-change-detection-on-levir-building-change-detection-test.md?metric=Precision>): Higher is better; unit: PERCENT.

- [Recall](<https://www.sota2.com/research/sota/building-change-detection-on-levir-building-change-detection-test.md?metric=Recall>): Higher is better; unit: PERCENT.

- [F1 Score](<https://www.sota2.com/research/sota/building-change-detection-on-levir-building-change-detection-test.md?metric=F1%20Score>): Higher is better; unit: PERCENT.

- [MCC](<https://www.sota2.com/research/sota/building-change-detection-on-levir-building-change-detection-test.md?metric=MCC>): Higher is better; unit: SCALAR.

- [mIoU](<https://www.sota2.com/research/sota/building-change-detection-on-levir-building-change-detection-test.md?metric=mIoU>): Higher is better; unit: PERCENT.

## Recorded results

| Method | Precision | Result date | Report paper |
| --- | --- | --- | --- |
| CEECNet V2 | 93.81 | Not recorded | 68fa6ac91e209131cea2a350 |
| CEECNet V1 | 93.73 | Not recorded | 68fa6ac91e209131cea2a350 |
| FracTAL ResNet | 93.63 | Not recorded | 68fa6ac91e209131cea2a350 |
| FracTAL ResNet | 93.6 | Not recorded | 68fa6ac91e209131cea2a350 |
| FracTAL ResNet | 93.5 | Not recorded | 68fa6ac91e209131cea2a350 |
| CEECNet V1 | 93.36 | Not recorded | 68fa6ac91e209131cea2a350 |
| Chen and Shi | 83.8 | Not recorded | 68fa6ac91e209131cea2a350 |

### CEECNet V2

Precision: 93.81. Result ID: 699ad0c65698665ec45fbb09.

[Reporting paper](<https://www.sota2.com/research/paper/68fa6ac91e209131cea2a350.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa6ac91e209131cea2a350.md>)

[Implementation](<https://github.com/feevos/ceecnet>)

Recorded evaluation configuration: {"depth (d)":5,"loss strategy":"evolving"}. This record does not assert independent reproduction.

### CEECNet V1

Precision: 93.73. Result ID: 699ad0c65698665ec45fbb08.

[Reporting paper](<https://www.sota2.com/research/paper/68fa6ac91e209131cea2a350.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa6ac91e209131cea2a350.md>)

[Implementation](<https://github.com/feevos/ceecnet>)

Recorded evaluation configuration: {"depth (d)":5,"loss strategy":"evolving"}. This record does not assert independent reproduction.

### FracTAL ResNet

Precision: 93.63. Result ID: 699ad0c65698665ec45fbb0c.

[Reporting paper](<https://www.sota2.com/research/paper/68fa6ac91e209131cea2a350.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa6ac91e209131cea2a350.md>)

[Implementation](<https://github.com/feevos/ceecnet>)

Recorded evaluation configuration: {"depth (d)":10,"loss strategy":"evolving"}. This record does not assert independent reproduction.

### FracTAL ResNet

Precision: 93.6. Result ID: 699ad0c65698665ec45fbb0b.

[Reporting paper](<https://www.sota2.com/research/paper/68fa6ac91e209131cea2a350.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa6ac91e209131cea2a350.md>)

[Implementation](<https://github.com/feevos/ceecnet>)

Recorded evaluation configuration: {"depth (d)":5,"loss strategy":"evolving"}. This record does not assert independent reproduction.

### FracTAL ResNet

Precision: 93.5. Result ID: 699ad0c65698665ec45fbb0a.

[Reporting paper](<https://www.sota2.com/research/paper/68fa6ac91e209131cea2a350.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa6ac91e209131cea2a350.md>)

[Implementation](<https://github.com/feevos/ceecnet>)

Recorded evaluation configuration: {"depth (d)":0,"loss strategy":"evolving"}. This record does not assert independent reproduction.

### CEECNet V1

Precision: 93.36. Result ID: 699ad0c65698665ec45fbb07.

[Reporting paper](<https://www.sota2.com/research/paper/68fa6ac91e209131cea2a350.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa6ac91e209131cea2a350.md>)

[Implementation](<https://github.com/feevos/ceecnet>)

Recorded evaluation configuration: {"depth (d)":5,"loss strategy":"static"}. This record does not assert independent reproduction.

### Chen and Shi

Precision: 83.8. Result ID: 699ad0c65698665ec45fbb06.

[Reporting paper](<https://www.sota2.com/research/paper/68fa6ac91e209131cea2a350.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.