Multiple Anchor Learning for Visual Object Detection

Benchmark Model Rank Results
object-detection-on-cocoMAL (ResNeXt101, multi-scale)#121box mAP: 47.0
object-detection-on-cocoMAL (ResNeXt101, single-scale)#133box mAP: 45.9
object-detection-on-cocoMAL (ResNet50, single-scale)#203box mAP: 39.2