# Change detection on Onera Satellite Change Detection (OSCD) (val)

Source: [Change detection on Onera Satellite Change Detection (OSCD) (val)](<https://www.sota2.com/research/sota/change-detection-on-onera-satellite-change-detection-oscd-val>)

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

Data updated: 2026-02-26T05:58:17.933Z

Evaluation of Change detection performance on the validation split of the Onera Satellite Change Detection (OSCD) dataset, reporting Precision, Recall, and F1 scores.

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 | Change detection |
| Dataset (including any stated split) | Onera Satellite Change Detection (OSCD) (val) |
| Metric | Precision |
| 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

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

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

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

## Recorded results

| Method | Precision | Result date | Report paper |
| --- | --- | --- | --- |
| Random init. | 70.53 | Not recorded | 68fa69eb1e209131cea24e50 |
| Imagenet (sup.) | 70.42 | Not recorded | 68fa69eb1e209131cea24e50 |
| MoCo-v2+TP | 69.14 | Not recorded | 68fa69eb1e209131cea24e50 |
| SeCo | 65.47 | Not recorded | 68fa69eb1e209131cea24e50 |
| MoCo-v2 | 64.49 | Not recorded | 68fa69eb1e209131cea24e50 |

### Random init.

Precision: 70.53. Result ID: 698df341ffc3901de099b95a.

[Reporting paper](<https://www.sota2.com/research/paper/68fa69eb1e209131cea24e50.md>)

Recorded evaluation configuration: {"Pre-training":"Random initialization","Backbone":"ResNet-18","Decoder":"U-Net","Evaluation Protocol":"Fine-tuning"}. This record does not assert independent reproduction.

### Imagenet (sup.)

Precision: 70.42. Result ID: 698df341ffc3901de099b95b.

[Reporting paper](<https://www.sota2.com/research/paper/68fa69eb1e209131cea24e50.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa61b41e209131ce9f58d2.md>)

Recorded evaluation configuration: {"Pre-training":"ImageNet supervised","Backbone":"ResNet-18","Decoder":"U-Net","Evaluation Protocol":"Fine-tuning"}. This record does not assert independent reproduction.

### MoCo-v2+TP

Precision: 69.14. Result ID: 698df341ffc3901de099b95d.

[Reporting paper](<https://www.sota2.com/research/paper/68fa69eb1e209131cea24e50.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa6bf01e209131cea31076.md>)

[Implementation](<https://www.sota2.com/geography-aware-ssl.github.io>)

Recorded evaluation configuration: {"Pre-training":"MoCo-v2 + Temporal Positional embedding","Backbone":"ResNet-18","Decoder":"U-Net","Evaluation Protocol":"Fine-tuning"}. This record does not assert independent reproduction.

### SeCo

Precision: 65.47. Result ID: 698df341ffc3901de099b95e.

[Reporting paper](<https://www.sota2.com/research/paper/68fa69eb1e209131cea24e50.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa69eb1e209131cea24e50.md>)

[Implementation](<https://github.com/ElementAI/seasonal-contrast>)

Recorded evaluation configuration: {"Pre-training":"SeCo (Seasonal Contrast)","Backbone":"ResNet-18","Decoder":"U-Net","Evaluation Protocol":"Fine-tuning"}. This record does not assert independent reproduction.

### MoCo-v2

Precision: 64.49. Result ID: 698df341ffc3901de099b95c.

[Reporting paper](<https://www.sota2.com/research/paper/68fa69eb1e209131cea24e50.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa67371e209131cea14e30.md>)

Recorded evaluation configuration: {"Pre-training":"MoCo-v2","Backbone":"ResNet-18","Decoder":"U-Net","Evaluation Protocol":"Fine-tuning"}. This record does not assert independent reproduction.

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