# Class-specific diffusion models improve military object detection in a low-data domain

arXiv: https://arxiv.org/abs/2604.18076 · HTML: https://mlsanity.com/paper/2604.18076/

## Key facts

- Cited by 3 papers in the map (data through 14 Oct 2026).
- Placed in Image generation, text-to-image and editing, one of 14 directions in Diffusion models.
- Builds on 17 of its 32 references that are in the map.
- First posted on arXiv 20 Apr 2026.

## Placed in

- Diffusion models / [Image generation, text-to-image and editing](https://mlsanity.com/diffusion/image-generation-text-to-image-and-editing/)

## Builds on (most cited first, 17 of 17)

- [LoRA: Low-Rank Adaptation of Large Language Models](https://mlsanity.com/paper/2106.09685/) (2021)
- [DINOv2: Learning Robust Visual Features without Supervision](https://mlsanity.com/paper/2304.07193/) (2023)
- [Generative Adversarial Networks](https://mlsanity.com/paper/2203.00667/) (2022)
- [Adding Conditional Control to Text-to-Image Diffusion Models](https://mlsanity.com/paper/2302.05543/) (2023)
- [Grounding DINO: Marrying DINO with Grounded Pre-Training for Open-Set Object Detection](https://mlsanity.com/paper/2303.05499/) (2023)
- [Kimi-VL Technical Report](https://mlsanity.com/paper/2504.07491/) (2025)
- [DiffusionDet: Diffusion Model for Object Detection](https://mlsanity.com/paper/2211.09788/) (2022)
- [Synthetic Data from Diffusion Models Improves ImageNet Classification](https://mlsanity.com/paper/2304.08466/) (2023)
- [TIDE: A General Toolbox for Identifying Object Detection Errors](https://mlsanity.com/paper/2008.08115/) (2020)
- [RF-DETR: Neural Architecture Search for Real-Time Detection Transformers](https://mlsanity.com/paper/2511.09554/) (2025)
- [LooseControl: Lifting ControlNet for Generalized Depth Conditioning](https://mlsanity.com/paper/2312.03079/) (2023)
- [AeroGen: Enhancing Remote Sensing Object Detection with Diffusion-Driven Data Generation](https://mlsanity.com/paper/2411.15497/) (2024)
- [T-LoRA: Single Image Diffusion Model Customization Without Overfitting](https://mlsanity.com/paper/2507.05964/) (2025)
- [Conditioning Diffusion Models via Attributes and Semantic Masks for Face Generation](https://mlsanity.com/paper/2306.00914/) (2023)
- [Implicit Multi-Spectral Transformer: An Lightweight and Effective Visible to Infrared Image Translation Model](https://mlsanity.com/paper/2404.07072/) (2024)
- [Improving Object Detector Training on Synthetic Data by Starting With a Strong Baseline Methodology](https://mlsanity.com/paper/2405.19822/) (2024)
- [Unlocking Thermal Aerial Imaging: Synthetic Enhancement of UAV Datasets](https://mlsanity.com/paper/2507.06797/) (2025)

## Built on by (most cited first, 3 of 3)

- [Training with synthetic data for drone detection in thermal imagery](https://mlsanity.com/paper/2608.17799/) (2026)
- [Domain shift-robust object detection with GenAI image editing](https://mlsanity.com/paper/2609.02299/) (2026)
- [Domain generalization and synthetic data in object detection: the enabler, the probe, and the gap](https://mlsanity.com/paper/2610.00030/) (2026)

## Abstract (arXiv)

Diffusion-based image synthesis has emerged as a promising source of synthetic training data for AI-based object detection and classification. In this work, we investigate whether images generated with diffusion can improve military vehicle detection under low-data conditions. We fine-tuned the text-to-image diffusion model FLUX.1 [dev] using LoRA with only 8 or 24 real images per class across 15 vehicle categories, resulting in class-specific diffusion models, which were used to generate new samples from automatically generated text prompts. The same real images were used to fine-tune the RF-DETR detector for a 15-class object detection task. Synthetic datasets generated by the diffusion models were then used to further improve detector performance. Importantly, no additional real data was required, as the generative models leveraged the same limited training samples. FLUX-generated images improved detection performance, particularly in the low-data regime (up to +8.0% mAP50 with 8 real samples). To address the limited geometric control of text prompt-based diffusion, we additionally generated structurally guided synthetic data using ControlNet with Canny edge-map conditioning, yielding a FLUX-ControlNet (FLUX-CN) dataset with explicit control over viewpoint and pose. Structural guidance further enhanced performance when data is scarce (+4.1% mAP50 with 8 real samples), but no additional benefit was observed when more real data is available. This study demonstrates that object-specific diffusion models are effective for improving military object detection in a low-data domain, and that structural guidance is most beneficial when real data is highly limited. These results highlight generative image data as an alternative to traditional simulation pipelines for the training of military AI systems.