RelationNet++: Bridging Visual Representations for Object Detection via Transformer Decoder

Introduction

@inproceedings{relationnetplusplus2020,
  title={RelationNet++: Bridging Visual Representations for Object Detection via Transformer Decoder},
  author={Chi, Cheng and Wei, Fangyun and Hu, Han},
  booktitle={NeurIPS},
  year={2020}
}

Results and models

Method	Backbone	MS train	Lr schd	box AP	Config	Download
RetinaNet+BVR	R-50	N	1x	38.5	config	Google
RetinaNet+BVR	X-101-DCN	Y	20e	46.5	config	Google
FCOS+BVR	X-101-DCN	Y	20e	48.9	config	Google
ATSS+BVR	X-101-DCN	Y	20e	50.7	config	Google