Skip to main content

> ML_BENCHMARK // COCO-OBJECT-DETECTION-MAP_v1.0

MS COCO Object Detection (AP @ IoU=0.50:0.95)

Object Detection · task-object-detection · active-headroom

Object Detectionactive-headroomhigh-vram-dependent
Official LeaderboardBack to Benchmarks

Evaluation Protocol

Evaluated on 5,000 COCO val2017 images or blind test-dev server averaging AP across 10 IoU thresholds from 0.50 to 0.95 with step 0.05.

Baseline & Metrics

Canonical Baseline:Faster R-CNN: 42.0 mAP | YOLOv8x: 53.9 mAP | Co-DETR + Swin-L: 66.0 mAP
Evaluated Metrics:
mean Average Precision (mAP 0.50:0.95)AP50AP75AP_smallAP_mediumAP_large

Contamination & Leakage Risks

Official blind test-dev server prevents test set ground truth leaks.

Reproducibility Concerns

Input image resolution and multi-scale testing flags must be explicitly specified.

CONNECTED DATASETMS COCO 2017 (Common Objects in Context)Computer Vision & Scene Understanding · 118,287 training images, 80 object categories, 5,000 validation images
View Dataset Spec