Open Vocabulary Object Detection with Proposal Mining and Prediction Equalization

Open paper

Benchmark	Model	Rank	Results
open-vocabulary-object-detection-on-lvis-v1-0	MEDet	#17	AP novel-LVIS base training: 22.4
open-vocabulary-object-detection-on-mscoco	MEDet (RN50)	#20	AP 0.5: 32.6