D-FINE: Redefine Regression Task in DETRs as Fine-grained Distribution Refinement

Benchmark Model Rank Results
real-time-object-detection-on-coco-val2017D-FINE-X+ (Objects365 pre-training)#4box AP: 59.3FPS (batch 1, GPU in brackets): 78 (T4)
real-time-object-detection-on-coco-val2017D-FINE-L+ (Objects365 pre-training)#10box AP: 57.1FPS (batch 1, GPU in brackets): 124 (T4)
real-time-object-detection-on-coco-val2017D-FINE-X#18box AP: 55.8FPS (batch 1, GPU in brackets): 78 (T4)
real-time-object-detection-on-coco-val2017D-FINE-M+ (Objects365 pre-training)#21box AP: 55.1FPS (batch 1, GPU in brackets): 178 (T4)
real-time-object-detection-on-coco-val2017D-FINE-L#33box AP: 54.0FPS (batch 1, GPU in brackets): 124 (T4)
real-time-object-detection-on-coco-val2017D-FINE-M#65box AP: 52.3FPS (batch 1, GPU in brackets): 180 (T4)
real-time-object-detection-on-coco-val2017D-FINE-S+ (Objects365 pre-training)#83box AP: 50.7FPS (batch 1, GPU in brackets): 287 (T4)
real-time-object-detection-on-coco-val2017D-FINE-S#100box AP: 48.5FPS (batch 1, GPU in brackets): 287 (T4)
object-detection-on-coco-val2017D-FINE-X (pretrained on Objects365) (extra data)#38box AP: 59.3
object-detection-on-coco-val2017D-FINE-L (pretrained on Objects365)#71box AP: 57.3
object-detection-on-coco-val2017D-FINE-X#95box AP: 55.8
object-detection-on-coco-val2017D-FINE-M (pretrained on Objects365)#107box AP: 55.1
object-detection-on-coco-val2017D-FINE-L#131box AP: 54.0
object-detection-on-coco-val2017D-FINE-M#171box AP: 52.3
object-detection-on-coco-val2017D-FINE-S (pretrained on Objects365)#202box AP: 50.7
object-detection-on-coco-val2017D-FINE-S#256box AP: 48.5
object-detection-on-coco-val2017D-FINE-N#353box AP: 42.8