# NAS-FPN: Learning Scalable Feature Pyramid Architecture for Object Detection

arXiv: https://arxiv.org/abs/1904.07392 · HTML: https://mlsanity.com/paper/1904.07392/

## Key facts

- Cited by 291 papers in the map (data through 14 Oct 2026).
- First posted on arXiv 16 Apr 2019.

## Built on by (most cited first, 30 of 291)

- [GShard: Scaling Giant Models with Conditional Computation and Automatic Sharding](https://mlsanity.com/paper/2006.16668/) (2020)
- [Swin Transformer V2: Scaling Up Capacity and Resolution](https://mlsanity.com/paper/2111.09883/) (2021)
- [YOLOv7: Trainable bag-of-freebies sets new state-of-the-art for real-time object detectors](https://mlsanity.com/paper/2207.02696/) (2022)
- [BEVDet: High-performance Multi-camera 3D Object Detection in Bird-Eye-View](https://mlsanity.com/paper/2112.11790/) (2021)
- [Mixture-of-Experts with Expert Choice Routing](https://mlsanity.com/paper/2202.09368/) (2022)
- [DETRs Beat YOLOs on Real-time Object Detection](https://mlsanity.com/paper/2304.08069/) (2023)
- [MViTv2: Improved Multiscale Vision Transformers for Classification and Detection](https://mlsanity.com/paper/2112.01526/) (2021)
- [Bottleneck Transformers for Visual Recognition](https://mlsanity.com/paper/2101.11605/) (2021)
- [YOLOv6: A Single-Stage Object Detection Framework for Industrial Applications](https://mlsanity.com/paper/2209.02976/) (2022)
- [RTMDet: An Empirical Study of Designing Real-Time Object Detectors](https://mlsanity.com/paper/2212.07784/) (2022)
- [Neural Architecture Search: Insights from 1000 Papers](https://mlsanity.com/paper/2301.08727/) (2023)
- [Transformer-Based Visual Segmentation: A Survey](https://mlsanity.com/paper/2304.09854/) (2023)
- [Towards Large-Scale Small Object Detection: Survey and Benchmarks](https://mlsanity.com/paper/2207.14096/) (2022)
- [You Only Look One-level Feature](https://mlsanity.com/paper/2103.09460/) (2021)
- [FlashOcc: Fast and Memory-Efficient Occupancy Prediction via Channel-to-Height Plugin](https://mlsanity.com/paper/2311.12058/) (2023)
- [A Survey of Modern Deep Learning based Object Detection Models](https://mlsanity.com/paper/2104.11892/) (2021)
- [CAT: Cross Attention in Vision Transformer](https://mlsanity.com/paper/2106.05786/) (2021)
- [Sample and Computation Redistribution for Efficient Face Detection](https://mlsanity.com/paper/2105.04714/) (2021)
- [TPH-YOLOv5: Improved YOLOv5 Based on Transformer Prediction Head for Object Detection on Drone-captured Scenarios](https://mlsanity.com/paper/2108.11539/) (2021)
- [LightTrack: Finding Lightweight Neural Networks for Object Tracking via One-Shot Architecture Search](https://mlsanity.com/paper/2104.14545/) (2021)
- [RigNet: Repetitive Image Guided Network for Depth Completion](https://mlsanity.com/paper/2107.13802/) (2021)
- [CBNet: A Composite Backbone Network Architecture for Object Detection](https://mlsanity.com/paper/2107.00420/) (2021)
- [RF-DETR: Neural Architecture Search for Real-Time Detection Transformers](https://mlsanity.com/paper/2511.09554/) (2025)
- [PAN++: Towards Efficient and Accurate End-to-End Spotting of Arbitrarily-Shaped Text](https://mlsanity.com/paper/2105.00405/) (2021)
- [Are we ready for a new paradigm shift? A Survey on Visual Deep MLP](https://mlsanity.com/paper/2111.04060/) (2021)
- [A Comprehensive Survey on Hardware-Aware Neural Architecture Search](https://mlsanity.com/paper/2101.09336/) (2021)
- [YOLOv6 v3.0: A Full-Scale Reloading](https://mlsanity.com/paper/2301.05586/) (2023)
- [HiSplat: Hierarchical 3D Gaussian Splatting for Generalizable Sparse-View Reconstruction](https://mlsanity.com/paper/2410.06245/) (2024)
- [Deep RGB-D Saliency Detection with Depth-Sensitive Attention and Automatic Multi-Modal Fusion](https://mlsanity.com/paper/2103.11832/) (2021)
- [Gold-YOLO: Efficient Object Detector via Gather-and-Distribute Mechanism](https://mlsanity.com/paper/2309.11331/) (2023)

## Abstract (arXiv)

Current state-of-the-art convolutional architectures for object detection are manually designed. Here we aim to learn a better architecture of feature pyramid network for object detection. We adopt Neural Architecture Search and discover a new feature pyramid architecture in a novel scalable search space covering all cross-scale connections. The discovered architecture, named NAS-FPN, consists of a combination of top-down and bottom-up connections to fuse features across scales. NAS-FPN, combined with various backbone models in the RetinaNet framework, achieves better accuracy and latency tradeoff compared to state-of-the-art object detection models. NAS-FPN improves mobile detection accuracy by 2 AP compared to state-of-the-art SSDLite with MobileNetV2 model in [32] and achieves 48.3 AP which surpasses Mask R-CNN [10] detection accuracy with less computation time.