# Object Detection on SODA-A Source: [Object Detection on SODA-A]() Collection index: [llms.txt]() Data updated: 2026-04-24T08:14:43.634Z Evaluation of Object Detection performance on the SODA-A dataset, reporting standard Average Precision metrics alongside specialized metrics (e.g., AI, HE, SV). 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) | SODA-A | | Metric | AI | | 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 - [AI](): Higher is better; unit: PERCENT. - [HE](): Higher is better; unit: PERCENT. - [SV](): Higher is better; unit: PERCENT. - [LV](): Higher is better; unit: PERCENT. - [SH](): Higher is better; unit: PERCENT. - [CO](): Higher is better; unit: PERCENT. - [ST](): Higher is better; unit: PERCENT. - [SP](): Higher is better; unit: PERCENT. - [WI](): Higher is better; unit: PERCENT. - [mAP](): Higher is better; unit: PERCENT. - [APs (Small)](): Higher is better; unit: PERCENT. - [APm (Medium)](): Higher is better; unit: PERCENT. - [APl (Large)](): Higher is better; unit: PERCENT. - [Latency (s)](): Lower is better; unit: TIME. - [Memory (GB)](): Lower is better; unit: SIZE. ## Recorded results | Method | AI | Result date | Report paper | | --- | --- | --- | --- | | UHR-DETR | 83.8 | Not recorded | 69eaf92f55a3c32af89f8a3c | | CEASC | 81.6 | Not recorded | 69eaf92f55a3c32af89f8a3c | | SparseFormer | 80 | Not recorded | 69eaf92f55a3c32af89f8a3c | | GigaDet | 68.8 | Not recorded | 69eaf92f55a3c32af89f8a3c | | SPDet | 64.5 | Not recorded | 69eaf92f55a3c32af89f8a3c | | RT-DETR | 29 | Not recorded | 69eaf92f55a3c32af89f8a3c | | Fovea | 5 | Not recorded | 69eaf92f55a3c32af89f8a3c | ### UHR-DETR AI: 83.8. Result ID: 69eb266047f34ce924d34842. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Backbone":"ResNet-50"}. This record does not assert independent reproduction. ### CEASC AI: 81.6. Result ID: 69eb266047f34ce924d3483f. [Reporting paper]() [Method origin paper]() [Implementation]() Recorded evaluation configuration: {"Backbone":"ResNet-50"}. This record does not assert independent reproduction. ### SparseFormer AI: 80. Result ID: 69eb266047f34ce924d3483e. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Backbone":"SparseNet","Training protocol":"from scratch for 36 epochs following \[34\]"}. This record does not assert independent reproduction. ### GigaDet AI: 68.8. Result ID: 69eb266047f34ce924d34841. [Reporting paper]() Recorded evaluation configuration: {"Backbone":"ResNet-50"}. This record does not assert independent reproduction. ### SPDet AI: 64.5. Result ID: 69eb266047f34ce924d34840. [Reporting paper]() Recorded evaluation configuration: {"Backbone":"CSPDarknet"}. This record does not assert independent reproduction. ### RT-DETR AI: 29. Result ID: 69eb266047f34ce924d3483c. [Reporting paper]() [Method origin paper]() [Implementation]() Recorded evaluation configuration: {"Backbone":"ResNet-50","Downsampling":"1024x1024"}. This record does not assert independent reproduction. ### Fovea AI: 5. Result ID: 69eb266047f34ce924d3483d. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Backbone":"ResNet-50","Downsampling":"1024x1024"}. This record does not assert independent reproduction. Page 1 of 1. This is an index page, not the full collection.