Stable Diffusion For Aerial Object Detection

TL;DR

This paper introduces a novel synthetic data augmentation framework for aerial object detection, leveraging diffusion models, ROI extraction, low-rank adaptation, and copy-paste techniques to improve detection accuracy.

Contribution

It presents a tailored augmentation framework that adapts diffusion models for aerial images, addressing semantic gaps and reducing retraining needs.

Findings

Enhanced detection performance with synthetic data.

Effective ROI extraction for sparse aerial objects.

Reduced retraining through low-rank adaptation.

Abstract

Aerial object detection is a challenging task, in which one major obstacle lies in the limitations of large-scale data collection and the long-tail distribution of certain classes. Synthetic data offers a promising solution, especially with recent advances in diffusion-based methods like stable diffusion (SD). However, the direct application of diffusion methods to aerial domains poses unique challenges: stable diffusion's optimization for rich ground-level semantics doesn't align with the sparse nature of aerial objects, and the extraction of post-synthesis object coordinates remains problematic. To address these challenges, we introduce a synthetic data augmentation framework tailored for aerial images. It encompasses sparse-to-dense region of interest (ROI) extraction to bridge the semantic gap, fine-tuning the diffusion model with low-rank adaptation (LORA) to circumvent exhaustive retraining, and finally, a Copy-Paste method to compose synthesized objects with backgrounds, providing a nuanced approach to aerial object detection through synthetic data.