# Object Detection on SA-Det-100k

Source: [Object Detection on SA-Det-100k](<https://www.sota2.com/research/sota/object-detection-on-sa-det-100k>)

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

Data updated: 2026-02-26T05:23:25.863Z

Evaluation of Object Detection performance on the SA-Det-100k dataset, reporting standard COCO-style AP metrics (AP@50:95, AP@50, AP@75, APs, APm, APL).

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) | SA-Det-100k |
| Metric | AP@50:95 |
| Unit | PERCENT |
| Direction | Higher is better |
| Results loaded | 2 of 2 |
| 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

- [AP@50:95](<https://www.sota2.com/research/sota/object-detection-on-sa-det-100k.md?metric=AP%4050%3A95>): Higher is better; unit: PERCENT.

- [AP@50](<https://www.sota2.com/research/sota/object-detection-on-sa-det-100k.md?metric=AP%4050>): Higher is better; unit: PERCENT.

- [AP@75](<https://www.sota2.com/research/sota/object-detection-on-sa-det-100k.md?metric=AP%4075>): Higher is better; unit: PERCENT.

- [APs](<https://www.sota2.com/research/sota/object-detection-on-sa-det-100k.md?metric=APs>): Higher is better; unit: PERCENT.

- [APm](<https://www.sota2.com/research/sota/object-detection-on-sa-det-100k.md?metric=APm>): Higher is better; unit: PERCENT.

- [APL](<https://www.sota2.com/research/sota/object-detection-on-sa-det-100k.md?metric=APL>): Higher is better; unit: PERCENT.

## Recorded results

| Method | AP@50:95 | Result date | Report paper |
| --- | --- | --- | --- |
| Relation-DETR | 45 | Not recorded | 68fa74201e209131cea5fcb4 |
| DINO with VFL | 43.7 | Not recorded | 68fa74201e209131cea5fcb4 |

### Relation-DETR

AP@50:95: 45. Result ID: 699330c13a6c314849145478.

[Reporting paper](<https://www.sota2.com/research/paper/68fa74201e209131cea5fcb4.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa74201e209131cea5fcb4.md>)

[Implementation](<https://github.com/xiuqhou/Relation-DETR>)

Recorded evaluation configuration: {"Backbone":"ResNet-50","Training schedule":"1x"}. This record does not assert independent reproduction.

### DINO with VFL

AP@50:95: 43.7. Result ID: 699330c13a6c314849145477.

[Reporting paper](<https://www.sota2.com/research/paper/68fa74201e209131cea5fcb4.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa6cf41e209131cea36cf4.md>)

[Implementation](<https://github.com/IDEACVR/DINO>)

Recorded evaluation configuration: {"Backbone":"ResNet-50","Training schedule":"1x"}. This record does not assert independent reproduction.

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