Parallel Residual Bi-Fusion Feature Pyramid Network for Accurate Single-Shot Object Detection

TL;DR

This paper introduces PRB-FPN, a novel feature pyramid network with bi-directional fusion and residual modules, significantly improving accuracy in single-shot object detection across various object sizes.

Contribution

The paper proposes a parallel bi-directional feature pyramid with residual and purification modules, enhancing localization and detection accuracy for small and large objects.

Findings

Achieves state-of-the-art results on UAVDT17 and MS COCO datasets.

Effectively detects both small and large objects simultaneously.

Facilitates easy training with deeper or lighter backbones.

Abstract

This paper proposes the Parallel Residual Bi-Fusion Feature Pyramid Network (PRB-FPN) for fast and accurate single-shot object detection. Feature Pyramid (FP) is widely used in recent visual detection, however the top-down pathway of FP cannot preserve accurate localization due to pooling shifting. The advantage of FP is weakened as deeper backbones with more layers are used. In addition, it cannot keep up accurate detection of both small and large objects at the same time. To address these issues, we propose a new parallel FP structure with bi-directional (top-down and bottom-up) fusion and associated improvements to retain high-quality features for accurate localization. We provide the following design improvements: (1) A parallel bifusion FP structure with a bottom-up fusion module (BFM) to detect both small and large objects at once with high accuracy. (2) A concatenation and re-organization (CORE) module provides a bottom-up pathway for feature fusion, which leads to the bi-directional fusion FP that can recover lost information from lower-layer feature maps. (3) The CORE feature is further purified to retain richer contextual information. Such CORE purification in both top-down and bottom-up pathways can be finished in only a few iterations. (4) The adding of a residual design to CORE leads to a new Re-CORE module that enables easy training and integration with a wide range of deeper or lighter backbones. The proposed network achieves state-of-the-art performance on the UAVDT17 and MS COCO datasets. Code is available at https://github.com/pingyang1117/PRBNet_PyTorch.