CBNet

2000년 도입 · 논문 3편에서 사용

CBNet is a backbone architecture that consists of multiple identical backbones (specially called Assistant Backbones and Lead Backbone) and composite connections between neighbor backbones. From left to right, the output of each stage in an Assistant Backbone, namely higher-level features, flows to the parallel stage of the succeeding backbone as part of inputs through composite connections. Finally, the feature maps of the last backbone named Lead Backbone are used for object detection. The features extracted by CBNet for object detection fuse the high-level and low-level features of multiple backbones, hence improve the detection performance.

출처: CBNet: A Novel Composite Backbone Network Architecture for Object Detection

소개 논문: CBNet: A Novel Composite Backbone Network Architecture for Object Detection