(ACM MM-2026) ADGNet: Asymmetric Dual-text Guided Network for Infrared Small Target Detection

Tongtong Wang¹, Mingzhu Xu^1✉, Chenglong Yu¹, Jing Wang¹, Xiaohui Lin¹, Weili Guan²

¹School of Software, Shandong University ²Harbin Institute of Technology, Shenzhen
^✉ Corresponding author

📢 Updates

• 🤗 [08/2026] Model checkpoints of the SOTA methods and text data are released on Hugging Face.
• 💻 [08/2026]Source code is now publicly available.
• 🎉 [07/2026] Our paper was accepted by ACM Multimedia 2026 (ACM MM 2026).

📖 Introduction

This repository provides the official implementation of ADGNet: Asymmetric Dual-text Guided Network for Infrared Small Target Detection, accepted by ACM Multimedia 2026 (ACM MM 2026).

Infrared Small Target Detection (IRSTD) aims to accurately segment weak and tiny targets from complex infrared backgrounds. Existing pure-vision methods rely solely on pixel-level information and often struggle to distinguish small targets from background clutter. Meanwhile, existing vision-language methods typically describe targets and backgrounds using a single textual prompt, overlooking their inherent semantic asymmetry and introducing feature optimization conflicts.

To address these limitations, we propose ADGNet, an Asymmetric Dual-text Guided Network for infrared small target detection. Its main components include:

• Asymmetric Dual-text Prompt (ADP): employs an abstract, image-independent target prompt and a detailed, image-dependent background prompt to provide dedicated semantic guidance.
• Asymmetric Dual-Branch Interaction (ADBI): separately constructs a Target Localization branch and a Background Suppression branch to enhance weak targets and suppress complex clutter.
• Adaptive Feature Aggregation (AFA): dynamically integrates the features produced by the two branches to achieve accurate target segmentation.

We also construct the Asymmetric Image-Text Infrared (AITIR) dataset by providing asymmetric text annotations for three widely used infrared small target datasets: IRSTD-1K, NUDT-SIRST, and SIRST. Extensive experiments demonstrate that ADGNet achieves competitive performance against 21 state-of-the-art methods.

This repository provides:

• 💻 Training and inference code for ADGNet
• 🏆 Model checkpoints of the SOTA methods
• 📝 AITIR asymmetric text annotations
• 📊 Evaluation scripts

🧠 Method / Framework

Figure 1. Overall architecture of ADGNet.

---

📁 Project Structure

ADGNet
├── Figs/# Overall framework of ADGNet
├── datasets/
│ ├── IRSTD-1K/
│ │ ├── images/# Original infrared images
│ │ ├── masks/# Ground-truth masks
│ │ ├── img_idx/# Training and testing splits
│ │ └── text/# Target and background prompts
│ ├── NUDT-SIRST/ │ └── SIRST/
├── model/
│ ├── ADGNet.py# Main ADGNet architecture
│ ├── TL_Branch.py# Target Localization branch
│ ├── BS_Branch.py# Background Suppression branch
│ └── DualStreamFusion.py# Dual-branch interaction and AFA
│
├── dataset.py# Dataset loader and augmentation
├── train.py# Training, testing, and inference
├── net.py# Network and loss wrapper
├── loss.py# SoftIoU loss
├── metrics.py# IoU, Pd, and Fa evaluation metrics
├── roc.py# ROC and AUC evaluation
├── params.py# FLOPs, parameters, and FPS analysis
├── utils.py# Data processing and training utilities
├── requirements.txt# Python dependencies
├── README.md
└── LICENSE

⚙️ Installation

📥 1. Clone the Repository

gitclonehttps://github.com/iLearn-Lab/MM26-ADGNet.gitcdMM26-ADGNet

🧪 2. Create the Environment

condacreate-nADGNetpython=3.10-ycondaactivateADGNet

🤗 3. Download the CLIP Weights

ADGNet uses the pretrained CLIP ViT-B/16 model. You can download the model weights from either Hugging Face or ModelScope.

sudoaptupdatesudoaptinstallgit-lfsgitlfsinstall# Option A: Download from Hugging Facegitclonehttps://huggingface.co/openai/clip-vit-base-patch16# Option B: Download from ModelScopegitclonehttps://www.modelscope.cn/openai-mirror/clip-vit-base-patch16.git

📦 4. Install Project Dependencies

pipinstalltorch==2.1.0torchvision==0.16.0torchaudio==2.1.0--index-urlhttps://download.pytorch.org/whl/cu121pipinstall-rrequirements.txt

🏆 Checkpoints / Models

We provide the trained checkpoints of ADGNet on three infrared small target detection datasets.

Dataset	IoU (%) ↑	Pd (%) ↑	Fa (10⁻⁶) ↓	Checkpoint
IRSTD-1K	72.38	93.20	4.10	Download
NUDT-SIRST	95.53	99.47	2.64	Download
SIRST	83.08	100.00	4.97	Download

📝 Text Annotations

We release the asymmetric text annotations used to construct the AITIR dataset.

Dataset	Text Annotations	Download
IRSTD-1K	Fixed Target Prompt + Detailed Background Prompt	Download
NUDT-SIRST
SIRST

🚀 Usage

🏋️ Training

Train ADGNet on the selected dataset:

pythontrain.py \
--trainset"IRSTD-1K" \
--testset"IRSTD-1K" \
--dataset_dir"./datasets" \
--epochs600 \
--batchSize16 \
--num_workers8 \
--modetrain

Replace IRSTD-1K with NUDT-SIRST or SIRST when training on another dataset.

🔍 Evaluation and Inference

Evaluate a trained checkpoint and generate prediction maps:

pythontrain.py \
--trainset"IRSTD-1K" \
--testset"IRSTD-1K" \
--dataset_dir"./datasets" \
--modetest \
--ckpt"./SOTA_pth/ADGNet_mIoU_72.38_IRSTD-1K.pth.tar"

🖼️ Visualization

The following figure presents qualitative comparisons between ADGNet and representative SOTA infrared small target detection methods on IRSTD-1K, NUDT-SIRST, and SIRST.

Figure 2. Qualitative comparisons of different methods on the IRSTD-1K, NUDT-SIRST, and SIRST datasets.

📚 Citation

If you find this project useful in your research, please consider citing our paper:

@inproceedings{wang2026adgnet,
title = {ADGNet: Asymmetric Dual-text Guided Network for Infrared Small Target Detection},
author = {Wang, Tongtong and Xu, Mingzhu and Yu, Chenglong and Wang, Jing and Lin, Xiaohui and Guan, Weili},
booktitle = {Proceedings of the ACM International Conference on Multimedia},
year = {2026},
}

Please also consider checking out and citing our other related work:

@inproceedings{yu2026dgnet,
title = {DGNet: Dual-knowledge Guided Network for Infrared Small Target Detection},
author = {Yu, Chenglong and Xu, Mingzhu and Wang, Jing and Wang, Tongtong and Miao, Pingping and Nie, Liqiang},
booktitle = {Proceedings of the ACM International Conference on Multimedia},
year = {2026},
}
@article{11017756,
author = {Xu, Mingzhu and Yu, Chenglong and Li, Zexuan and Tang, Haoyu and Hu, Yupeng and Nie, Liqiang},
journal = {IEEE Transactions on Geoscience and Remote Sensing},
title = {HDNet: A Hybrid Domain Network With Multiscale High-Frequency Information Enhancement for Infrared Small-Target Detection},
year = {2025},
volume = {63},
pages = {1--15},
doi = {10.1109/TGRS.2025.3574962},
}

🛠️ IRSTD-AutoLabel

In addition, we have open-sourced an automated annotation tool for infrared small target detection, IRSTD-AutoLabel. Interested readers are encouraged to visit the project page for more details and usage instructions:

Project:：iLearn-Lab/IRSTD-AutoLabel

📄 License

This project is released under the Apache License 2.0.

You may use, modify, and distribute the code in accordance with the terms of the license. Please retain the original license and attribution notices in redistributed or modified versions.

📧 Contact

If you have any questions, please feel free to contact us at:wangtongtong@163.comorwangttong@mail.sdu.edu.cn.