Skip to main navigation Skip to search Skip to main content

EDSOD: An Encoder-Decoder, Diffusion-model, and Swin-Transformer-based Small Object Detector

Research output: Chapter in Book/Report/Conference proceedingConference contribution

Abstract

Small object detection (SOD) given aerial images suffers from an information imbalance across different feature scales. This makes it extremely challenging to perform accurate SOD. Existing methods, e.g., Feature Pyramid Network (FPN)-based algorithms, focus on extracting high-resolution and low-resolution semantic features from different convolution layers. However, in deeper convolution layers, semantic feature misalignment and the loss of key information are inevitable. To tackle such issues, this work proposes a new encoder-decoder-based SOD framework with a Diffusion Model and Swin Transformer given aerial images. First, we reformulate an SOD task as a Noise-to-Box process. We then construct an encoder-decoder-based framework by using a diffusion model and Swin Transformer for dynamic bounding box generation. We introduce a decoupling training and inferencing strategy to recognize and locate small objects accurately. We finally evaluate the proposed framework on several public benchmarks. The experimental results well show its better SOD performance than the state of the art. Code is available at https://github.com/BrainPotter/EDSOD.

Original languageEnglish (US)
Title of host publicationIROS 2025 - 2025 IEEE/RSJ International Conference on Intelligent Robots and Systems, Conference Proceedings
EditorsChristian Laugier, Alessandro Renzaglia, Nikolay Atanasov, Stan Birchfield, Grzegorz Cielniak, Leonardo De Mattos, Laura Fiorini, Philippe Giguere, Kenji Hashimoto, Javier Ibanez-Guzman, Tetsushi Kamegawa, Jinoh Lee, Giuseppe Loianno, Kevin Luck, Hisataka Maruyama, Philippe Martinet, Hadi Moradi, Urbano Nunes, Julien Pettre, Alberto Pretto, Tommaso Ranzani, Arne Ronnau, Silvia Rossi, Elliott Rouse, Fabio Ruggiero, Olivier Simonin, Danwei Wang, Ming Yang, Eiichi Yoshida, Huijing Zhao
PublisherInstitute of Electrical and Electronics Engineers Inc.
Pages1659-1665
Number of pages7
ISBN (Electronic)9798331543938
DOIs
StatePublished - 2025
Externally publishedYes
Event2025 IEEE/RSJ International Conference on Intelligent Robots and Systems, IROS 2025 - Hangzhou, China
Duration: Oct 19 2025Oct 25 2025

Publication series

NameIEEE International Conference on Intelligent Robots and Systems
ISSN (Print)2153-0858
ISSN (Electronic)2153-0866

Conference

Conference2025 IEEE/RSJ International Conference on Intelligent Robots and Systems, IROS 2025
Country/TerritoryChina
CityHangzhou
Period10/19/2510/25/25

All Science Journal Classification (ASJC) codes

  • Control and Systems Engineering
  • Software
  • Computer Vision and Pattern Recognition
  • Computer Science Applications

Keywords

  • Small object detection
  • Swin Transformer
  • aerial images
  • diffusion model
  • encoder-decoder framework

Fingerprint

Dive into the research topics of 'EDSOD: An Encoder-Decoder, Diffusion-model, and Swin-Transformer-based Small Object Detector'. Together they form a unique fingerprint.

Cite this