SIMPL: Generating Synthetic Overhead Imagery to Address Custom Zero-Shot and Few-Shot Detection Problems

Yang Xu, Bohao Huang, Xiong Luo, Kyle Bradbury, Jordan M. Malof

Research output: Contribution to journalArticlepeer-review

8 Scopus citations

Abstract

Recently deep neural networks (DNNs) have achieved tremendous success for object detection in overhead (e.g., satellite) imagery. One ongoing challenge however is the acquisition of training data, due to high costs of obtaining satellite imagery and annotating objects in it. In this article, we present a simple approach - termed Synthetic object IMPLantation (SIMPL) - to easily and rapidly generate large quantities of synthetic overhead training data for custom target objects. We demonstrate the effectiveness of using SIMPL synthetic imagery for training DNNs in zero-shot scenarios where no real imagery is available; and few-shot learning scenarios, where limited real-world imagery is available. We also conduct experiments to study the sensitivity of SIMPL's effectiveness to some key design parameters, providing users for insights when designing synthetic imagery for custom objects. We release a software implementation of our SIMPL approach, as well as design details of our experimental synthetic imagery, so that others can build upon our approach, or use it for their own custom problems.

Original languageEnglish
Pages (from-to)4386-4396
Number of pages11
JournalIEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing
Volume15
DOIs
StatePublished - 2022

Keywords

  • Few-shot
  • object detection
  • overhead imagery
  • synthetic data
  • zero-shot

Fingerprint

Dive into the research topics of 'SIMPL: Generating Synthetic Overhead Imagery to Address Custom Zero-Shot and Few-Shot Detection Problems'. Together they form a unique fingerprint.

Cite this