• Home
  • Search
  • SsFPN: Scale Sequence (S2) Feature-Based Feature Pyramid Network for Object Detection.
  • Cite Icon53
  • https://doi.org/10.3390/s23094432Copy DOI Icon

SsFPN: Scale Sequence (S2) Feature-Based Feature Pyramid Network for Object Detection.

Show More
  • Abstract
  • Highlights & Summary
  • Literature Map
  • References
  • Citations
  • Similar Papers
Abstract

Object detection is a fundamental task in computer vision. Over the past several years, convolutional neural network (CNN)-based object detection models have significantly improved detection accuracyin terms of average precision (AP). Furthermore, feature pyramid networks (FPNs) are essential modules for object detection models to consider various object scales. However, the AP for small objects is lower than the AP for medium and large objects. It is difficult to recognize small objects because they do not have sufficient information, and information is lost in deeper CNN layers. This paper proposes a new FPN model named ssFPN (scale sequence (S2) feature-based feature pyramid network) to detect multi-scale objects, especially small objects. We propose a new scale sequence (S2) feature that is extracted by 3D convolution on the level of the FPN. It is defined and extracted from the FPN to strengthen the information on small objects based on scale-space theory. Motivated by this theory, the FPN is regarded as a scale space and extracts a scale sequence (S2) feature by three-dimensional convolution on the level axis of the FPN. The defined feature is basically scale-invariant and is built on a high-resolution pyramid feature map for small objects. Additionally, the deigned S2 feature can be extended to most object detection models based on FPNs. We also designed a feature-level super-resolution approach to show the efficiency of the scale sequence (S2) feature. We verified that the scale sequence (S2) feature could improve the classification accuracy for low-resolution images by training a feature-level super-resolution model. To demonstrate the effect of the scale sequence (S2) feature, experiments on the scale sequence (S2) feature built-in object detection approach including both one-stage and two-stage models were conducted on the MS COCO dataset. For the two-stage object detection models Faster R-CNN and Mask R-CNN with the S2 feature, AP improvements of up to 1.6% and 1.4%, respectively, were achieved. Additionally, the APS of each model was improved by 1.2% and 1.1%, respectively. Furthermore, the one-stage object detection models in the YOLO series were improved. For YOLOv4-P5, YOLOv4-P6, YOLOR-P6, YOLOR-W6, and YOLOR-D6 with the S2 feature, 0.9%, 0.5%, 0.5%, 0.1%, and 0.1% AP improvements were observed. For small object detection, the APS increased by 1.1%, 1.1%, 0.9%, 0.4%, and 0.1%, respectively. Experiments using the feature-level super-resolution approach with the proposed scale sequence (S2) feature were conducted on the CIFAR-100 dataset. By training the feature-level super-resolution model, we verified that ResNet-101 with the S2 feature trained on LR images achieved a 55.2% classification accuracy, which was 1.6% higher than for ResNet-101 trained on HR images.

Similar Papers
  • Research Article
  • Citations48

Deer survey from drone thermal imagery using enhanced faster R-CNN based on ResNets and FPN

  • Nov 25, 2023
  • Ecological Informatics
  • Haitao Lyu +5
  • Research Article
  • Citations67

A novel deep learning‐based single shot multibox detector model for object detection in optical remote sensing images

  • May 16, 2022
  • Geoscience Data Journal
  • Liguo Wang +7
  • PDF
  • Research Article
  • Citations5

Bird Object Detection: Dataset Construction, Model Performance Evaluation, and Model Lightweighting.

  • Sep 14, 2023
  • Animals : an open access journal from MDPI
  • Yang Wang +6
  • Research Article
  • Citations8

Road Object Detection using SSD-MobileNet Algorithm: Case Study for Real-Time ADAS Applications

  • Mar 18, 2024
  • Journal of Robotics and Control (JRC)
  • Omar Bouazizi +3
  • PDF
  • Research Article
  • Citations5

Research on Object Detection Model Based on Feature Network Optimization

  • Sep 14, 2021
  • Processes
  • Xiaoliang Zhang +3
  • Book Chapter
  • Citations1

Ship Detection from Satellite Imagery Using RetinaNet with Instance Segmentation

  • Jan 01, 2023
  • Arya Dhorajiya +2
  • PDF
  • Research Article
  • Citations22

Detecting common coccinellids found in sorghum using deep learning models

  • Jun 16, 2023
  • Scientific Reports
  • Chaoxin Wang +3
  • PDF
  • Research Article
  • Citations10

Traffic Light Detection by Integrating Feature Fusion and Attention Mechanism

  • Sep 04, 2023
  • Electronics
  • Chi-Hung Chuang +3
  • Conference Article
  • Citations22

TFPN: Twin Feature Pyramid Networks for Object Detection

  • Nov 01, 2019
  • Yi Liang +6
  • Dissertation

Deep learning for small object detection in images

  • Jul 01, 2020
  • Yang Liu
  • Research Article

MambaFPN: A SSM-based feature pyramid network for object detection.

  • Jun 01, 2026
  • Neural networks : the official journal of the International Neural Network Society
  • Le Liang +2
  • Research Article
  • Citations2

Improvement in Object Detection Using Multi-Scale RoI Pooling and Feature Pyramid Network

  • Mar 31, 2022
  • Journal of Computing Science and Engineering
  • Seungtae Nam +1
  • PDF
  • Research Article
  • Citations60

Detection of Small Ship Objects Using Anchor Boxes Cluster and Feature Pyramid Network Model for SAR Imagery

  • Feb 12, 2020
  • Journal of Marine Science and Engineering
  • Peng Chen +4
  • Research Article
  • Citations9

Construct Effective Geometry Aware Feature Pyramid Network for Multi-Scale Object Detection

  • Jun 28, 2022
  • Proceedings of the AAAI Conference on Artificial Intelligence
  • Jinpeng Dong +4
  • Conference Article
  • Citations1

An Autonomous Garbage Removal Approach Using Resnet as Backbone and MaskRcnn for Segmentation

  • Aug 18, 2023
  • Sirapu Ashajyothi +1
Cactus Communications logo

Copyright 2026 Cactus Communications. All rights reserved.