CBNet: A Novel Composite Backbone Network Architecture for Object Detection

9 Sep 2019  ·  Yudong Liu, Yongtao Wang, Siwei Wang, Ting-Ting Liang, Qijie Zhao, Zhi Tang, Haibin Ling ·

In existing CNN based detectors, the backbone network is a very important component for basic feature extraction, and the performance of the detectors highly depends on it. In this paper, we aim to achieve better detection performance by building a more powerful backbone from existing backbones like ResNet and ResNeXt. Specifically, we propose a novel strategy for assembling multiple identical backbones by composite connections between the adjacent backbones, to form a more powerful backbone named Composite Backbone Network (CBNet). In this way, CBNet iteratively feeds the output features of the previous backbone, namely high-level features, as part of input features to the succeeding backbone, in a stage-by-stage fashion, and finally the feature maps of the last backbone (named Lead Backbone) are used for object detection. We show that CBNet can be very easily integrated into most state-of-the-art detectors and significantly improve their performances. For example, it boosts the mAP of FPN, Mask R-CNN and Cascade R-CNN on the COCO dataset by about 1.5 to 3.0 percent. Meanwhile, experimental results show that the instance segmentation results can also be improved. Specially, by simply integrating the proposed CBNet into the baseline detector Cascade Mask R-CNN, we achieve a new state-of-the-art result on COCO dataset (mAP of 53.3) with single model, which demonstrates great effectiveness of the proposed CBNet architecture. Code will be made available on

PDF Abstract
Task Dataset Model Metric Name Metric Value Global Rank Result Benchmark
Instance Segmentation COCO test-dev Cascade Mask R-CNN (ResNeXt152, CBNet) mask AP 43.3 # 47
Object Detection COCO test-dev Cascade Mask R-CNN (Triple-ResNeXt152, multi-scale) box mAP 53.3 # 59
AP50 71.9 # 19
AP75 58.5 # 22
APS 35.5 # 16
APM 55.8 # 23
APL 66.7 # 16
Hardware Burden None # 1
Operations per network pass None # 1