Object detection is the task of detecting instances of objects of a certain class within an image.
What will the value of the herein defined Object Detection Performance Index be on 2026-12-14?
The index is constructed as follows:
We take the average (arithmetic mean) of - ln (error) of the state-of-the-art performance across all benchmarks in the index
The index is then defined by scaling this mean so that its average value for the year 2019 is 100
The following benchmarks are included in the Object Detection Performance Index:
Object detection on: COCO test, COCO minival, CrowdHuman (full body). 3D object detection on: KITTI Cars Moderate, KITTI Cars Easy, KITTI Cars Hard, KITTI Cyclists Hard, KITTI Pedestrians Moderate, SUN-RGBD val, Real-time object detection on COCO, and Weakly Supervised object detection on Pascal VOC 2007.
Historical data on the Object Detection Performance Index may be found here.
This question resolves as the value of this index on 2026-12-14, 11:59PM GMT.
Models that are trained on multiple datasets do not qualify for the purpose of this question—only models trained on benchmark-specific datasets will be considered.
A benchmark will be removed from the index if:
- At the time of resolution no new performance data is available for new models for the specific benchmark over the previous 6 months
- The value of - ln (1 - error) for that benchmark exceeds 50
If a benchmark is removed from the index, the index shall simply be re-constructed according the procedure outlined above.
Performance figures may be taken from e-prints, conference papers, peer-reviewed articles, and blog articles by reputable AI labs (including the associated code repositories). Published performance figures must be available before 2026-12-14, 11:59PM GMT to qualify.
For the purpose of the index, error is calculated as 1-(average precision)/100.