Abstract:Feature pyramids are widely used to improve scale invariance for object detection. Most methods just map the objects to feature maps with relevant square receptive fields, but rarely pay attention to the aspect ratio variation, which is also an important property of object instances. It will lead to a poor match between rectangular objects and assigned features with square receptive fields, thus preventing from accurate recognition and location. Besides, the information propagation among feature layers is sparse, namely, each feature in the pyramid may mainly or only contain single-level information, which is not representative enough for classification and localization sub-tasks. In this paper, Bidirectional Matrix Feature Pyramid Network (BMFPN) is proposed to address these issues. It consists of three modules: Diagonal Layer Generation Module (DLGM), Top-down Module (TDM) and Bottom-up Module (BUM). First, multi-level features extracted by backbone are fed into DLGM to produce the base features. Then these base features are utilized to construct the final feature pyramid through TDM and BUM in series. The receptive fields of the designed feature layers in BMFPN have various scales and aspect ratios. Objects can be correctly assigned to appropriate and representative feature maps with relevant receptive fields depending on its scale and aspect ratio properties. Moreover, TDM and BUM form bidirectional and reticular information flow, which effectively fuses multi-level information in top-down and bottom-up manner respectively. To evaluate the effectiveness of our proposed architecture, an end-to-end anchor-free detector is designed and trained by integrating BMFPN into FCOS. And the center-ness branch in FCOS is modified with our Gaussian center-ness branch (GCB), which brings another slight improvement. Without bells and whistles, our method gains +3.3%, +2.4% and +2.6% AP on MS COCO dataset from baselines with ResNet-50, ResNet-101 and ResNeXt-101 backbones, respectively.

MM-FPN: Multi-path and Multi-scale Feature Pyramid Network for Object Detection

Multi-scale Convolution Target Detection Algorithm with Feature Pyramid

SAFPN: a Full Semantic Feature Pyramid Network for Object Detection

Enhanced semantic feature pyramid network for small object detection

MFPN: A Novel Mixture Feature Pyramid Network of Multiple Architectures for Object Detection

MGFPN: Enhancing Multi-Scale Feature for Object Detection

Multi-Scale Residual Aggregation Feature Pyramid Network for Object Detection

Dynamic Feature Pyramid Networks for Detection

Concise Feature Pyramid Region Proposal Network for Multi-Scale Object Detection.

Cascaded Multi-Channel Feature Fusion for Object Detection.

Extended Feature Pyramid Network for Small Object Detection

HA-FPN: Hierarchical Attention Feature Pyramid Network for Object Detection

Complementary Feature Pyramid Network for Object Detection

Learning Discriminated Features Based on Feature Pyramid Networks and Attention for Multi-scale Object Detection

Pyramid attention object detection network with multi-scale feature fusion

AFPN: Asymptotic Feature Pyramid Network for Object Detection

Improving Multiscale Object Detection With Off-Centered Semantics Refinement

Bidirectional Matrix Feature Pyramid Network for Object Detection

UEFPN: Unified and Enhanced Feature Pyramid Networks for Small Object Detection

Attention-guided Context Feature Pyramid Network for Object Detection

Bidirectional Parallel Feature Pyramid Network for Object Detection