# Object Detection on MLT (test)

Source: [Object Detection on MLT (test)](<https://www.sota2.com/research/sota/object-detection-on-mlt-test>)

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

Data updated: 2026-02-26T06:01:12.356Z

Evaluation of Object Detection performance on the test split of the MLT dataset, reporting standard detection metrics such as Precision, Recall, and H-mean.

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 Detection |
| Dataset (including any stated split) | MLT (test) |
| Metric | Recall |
| Unit | PERCENT |
| Direction | Higher is better |
| Results loaded | 3 of 3 |
| 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](<https://www.sota2.com/research/sota/object-detection-on-mlt-test.md?metric=Recall>): Higher is better; unit: PERCENT.

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

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

## Recorded results

| Method | Recall | Result date | Report paper |
| --- | --- | --- | --- |
| RetinaNet-BCL | 58.91 | Not recorded | 68fa6a091e209131cea25a5d |
| RetinaNet-CSL | 58.32 | Not recorded | 68fa6a091e209131cea25a5d |
| RetinaNet-Reg | 55.7 | Not recorded | 68fa6a091e209131cea25a5d |

### RetinaNet-BCL

Recall: 58.91. Result ID: 698e0235ffc3901de09a19ce.

[Reporting paper](<https://www.sota2.com/research/paper/68fa6a091e209131cea25a5d.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa6a091e209131cea25a5d.md>)

[Implementation](<https://github.com/Thinklab-SJTU/DCL_RetinaNet_Tensorflow>)

Recorded evaluation configuration: {"Backbone":"ResNet101","Data Augmentation":"true","Multi-scale training and testing":"true"}. This record does not assert independent reproduction.

### RetinaNet-CSL

Recall: 58.32. Result ID: 698e0235ffc3901de09a19cd.

[Reporting paper](<https://www.sota2.com/research/paper/68fa6a091e209131cea25a5d.md>)

Recorded evaluation configuration: {"Backbone":"ResNet101","Data Augmentation":"true","Multi-scale training and testing":"true"}. This record does not assert independent reproduction.

### RetinaNet-Reg

Recall: 55.7. Result ID: 698e0235ffc3901de09a19cc.

[Reporting paper](<https://www.sota2.com/research/paper/68fa6a091e209131cea25a5d.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa640c1e209131cea02556.md>)

[Implementation](<https://github.com/facebookresearch/Detectron>)

Recorded evaluation configuration: {"Backbone":"ResNet101","Data Augmentation":"true","Multi-scale training and testing":"true"}. This record does not assert independent reproduction.

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