# Dual Refinement Feature Pyramid Networks for Object Detection

arXiv: https://arxiv.org/abs/2012.01733 · HTML: https://mlsanity.com/paper/2012.01733/

## Key facts

- Cited by 8 papers in the map (data through 14 Oct 2026).
- First posted on arXiv 3 Dec 2020.

## Built on by (most cited first, 8 of 8)

- [AFPN: Asymptotic Feature Pyramid Network for Object Detection](https://mlsanity.com/paper/2306.15988/) (2023)
- [A DeNoising FPN With Transformer R-CNN for Tiny Object Detection](https://mlsanity.com/paper/2406.05755/) (2024)
- [Cross-Layer Feature Pyramid Transformer for Small Object Detection in Aerial Images](https://mlsanity.com/paper/2407.19696/) (2024)
- [CE-FPN: Enhancing Channel Information for Object Detection](https://mlsanity.com/paper/2103.10643/) (2021)
- [Butter: Frequency Consistency and Hierarchical Fusion for Autonomous Driving Object Detection](https://mlsanity.com/paper/2507.13373/) (2025)
- [PulseFi: A Low Cost Robust Machine Learning System for Accurate Cardiopulmonary and Apnea Monitoring Using Channel State Information](https://mlsanity.com/paper/2510.24744/) (2025)
- [Building Facade Parsing R-CNN](https://mlsanity.com/paper/2205.05912/) (2022)
- [MO R-CNN: Multispectral Oriented R-CNN for Object Detection in Remote Sensing Image](https://mlsanity.com/paper/2509.16957/) (2025)

## Abstract (arXiv)

FPN is a common component used in object detectors, it supplements multi-scale information by adjacent level features interpolation and summation. However, due to the existence of nonlinear operations and the convolutional layers with different output dimensions, the relationship between different levels is much more complex, the pixel-wise summation is not an efficient approach. In this paper, we first analyze the design defects from pixel level and feature map level. Then, we design a novel parameter-free feature pyramid networks named Dual Refinement Feature Pyramid Networks (DRFPN) for the problems. Specifically, DRFPN consists of two modules: Spatial Refinement Block (SRB) and Channel Refinement Block (CRB). SRB learns the location and content of sampling points based on contextual information between adjacent levels. CRB learns an adaptive channel merging method based on attention mechanism. Our proposed DRFPN can be easily plugged into existing FPN-based models. Without bells and whistles, for two-stage detectors, our model outperforms different FPN-based counterparts by 1.6 to 2.2 AP on the COCO detection benchmark, and 1.5 to 1.9 AP on the COCO segmentation benchmark. For one-stage detectors, DRFPN improves anchor-based RetinaNet by 1.9 AP and anchor-free FCOS by 1.3 AP when using ResNet50 as backbone. Extensive experiments verifies the robustness and generalization ability of DRFPN. The code will be made publicly available.