# Object Detection on SODA10M night

Source: [Object Detection on SODA10M night](<https://www.sota2.com/research/sota/object-detection-on-soda10m-night>)

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

Data updated: 2026-02-26T05:38:56.522Z

Evaluation of standard Object Detection performance on the 'night' subset of the SODA10M dataset, reporting class-specific metrics for Car, Pedestrian, and Cyclist.

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) | SODA10M night |
| Metric | Car AP |
| 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

- [Car AP](<https://www.sota2.com/research/sota/object-detection-on-soda10m-night.md?metric=Car%20AP>): Higher is better; unit: PERCENT.

- [Pedestrian AP](<https://www.sota2.com/research/sota/object-detection-on-soda10m-night.md?metric=Pedestrian%20AP>): Higher is better; unit: PERCENT.

- [Cyclist AP](<https://www.sota2.com/research/sota/object-detection-on-soda10m-night.md?metric=Cyclist%20AP>): Higher is better; unit: PERCENT.

## Recorded results

| Method | Car AP | Result date | Report paper |
| --- | --- | --- | --- |
| FS (fully supervised) | 86.9 | Not recorded | 68fa6c2a1e209131cea3277e |
| DUA | 77.2 | Not recorded | 68fa6c2a1e209131cea3277e |
| SO (source-only) | 75.3 | Not recorded | 68fa6c2a1e209131cea3277e |

### FS (fully supervised)

Car AP: 86.9. Result ID: 698f50673a6c314849f7f7ec.

[Reporting paper](<https://www.sota2.com/research/paper/68fa6c2a1e209131cea3277e.md>)

Recorded evaluation configuration: {"Architecture":"YOLOv3","Supervision":"fully supervised"}. This record does not assert independent reproduction.

### DUA

Car AP: 77.2. Result ID: 698f50673a6c314849f7f7eb.

[Reporting paper](<https://www.sota2.com/research/paper/68fa6c2a1e209131cea3277e.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa6c2a1e209131cea3277e.md>)

[Implementation](<https://github.com/jmiemirza/DUA>)

Recorded evaluation configuration: {"Architecture":"YOLOv3","Source Training Split":"SODA10M day","Adaptation Mode":"Dynamic Unsupervised Adaptation"}. This record does not assert independent reproduction.

### SO (source-only)

Car AP: 75.3. Result ID: 698f50673a6c314849f7f7ea.

[Reporting paper](<https://www.sota2.com/research/paper/68fa6c2a1e209131cea3277e.md>)

Recorded evaluation configuration: {"Architecture":"YOLOv3","Supervision":"source-only","Source Training Split":"SODA10M day"}. This record does not assert independent reproduction.

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