# Dynamic Feature Pyramid Networks for Object Detection

arXiv: https://arxiv.org/abs/2012.00779 · HTML: https://mlsanity.com/paper/2012.00779/

## Key facts

- Cited by 8 papers in the map (data through 14 Oct 2026).
- First posted on arXiv 1 Dec 2020.

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

- [DynamicDet: A Unified Dynamic Architecture for Object Detection](https://mlsanity.com/paper/2304.05552/) (2023)
- [FAQ: Feature Aggregated Queries for Transformer-based Video Object Detectors](https://mlsanity.com/paper/2303.08319/) (2023)
- [Image Captioning via Dynamic Path Customization](https://mlsanity.com/paper/2406.00334/) (2024)
- [Image super-resolution via dynamic network](https://mlsanity.com/paper/2310.10413/) (2023)
- [EPANet: Efficient Path Aggregation Network for Underwater Fish Detection](https://mlsanity.com/paper/2508.00528/) (2025)
- [Layout-Independent License Plate Recognition via Integrated Vision and Language Models](https://mlsanity.com/paper/2510.10533/) (2025)
- [Boundary and Position Information Mining for Aerial Small Object Detection](https://mlsanity.com/paper/2601.16617/) (2026)
- [FinSight-Net:A Physics-Aware Decoupled Network with Frequency-Domain Compensation for Underwater Fish Detection in Smart Aquaculture](https://mlsanity.com/paper/2602.19437/) (2026)

## Abstract (arXiv)

Feature pyramid network (FPN) is a critical component in modern object detection frameworks. The performance gain in most of the existing FPN variants is mainly attributed to the increase of computational burden. An attempt to enhance the FPN is enriching the spatial information by expanding the receptive fields, which is promising to largely improve the detection accuracy. In this paper, we first investigate how expanding the receptive fields affect the accuracy and computational costs of FPN. We explore a baseline model called inception FPN in which each lateral connection contains convolution filters with different kernel sizes. Moreover, we point out that not all objects need such a complicated calculation and propose a new dynamic FPN (DyFPN). The output features of DyFPN will be calculated by using the adaptively selected branch according to a dynamic gating operation. Therefore, the proposed method can provide a more efficient dynamic inference for achieving a better trade-off between accuracy and computational cost. Extensive experiments conducted on MS-COCO benchmark demonstrate that the proposed DyFPN significantly improves performance with the optimal allocation of computation resources. For instance, replacing inception FPN with DyFPN reduces about 40% of its FLOPs while maintaining similar high performance.