datasets
Training and evaluation data, with the modality, task and licence stated up front. Listed live from the Hugging Face Hub.
DisasterM3
DisasterM3: A Remote Sensing Vision-Language Dataset for Disaster Damage Assessment and Response
Junjue Wang*,
Weihao Xuan*,
Heli Qi, Zhihao Liu, Kunyi Liu, Yuhan Wu, Hongruixuan Chen,
Jian Song
Junshi Xia, Zhuo Zheng, Naoto Yokoya†
* Equal Contributions
† Corresponding Author
Paper: https://arxiv.org/abs/2505.21089
Code: https://github.com/Junjue-Wang/DisasterM3
Highlights
DisasterM3 includes 26,988 bi-temporal satellite images and 123k instruction pairs across… See the full description on the dataset page: https://huggingface.co/datasets/Kingdrone-Junjue/DisasterM3.DisasterM3
DisasterM3: A Remote Sensing Vision-Language Dataset for Disaster Damage Assessment and Response
Junjue Wang*,
Weihao Xuan*,
Heli Qi, Zhihao Liu, Kunyi Liu, Yuhan Wu, Hongruixuan Chen,
Jian Song
Junshi Xia, Zhuo Zheng, Naoto Yokoya†
* Equal Contributions
† Corresponding Author
Paper: https://arxiv.org/abs/2505.21089
Code: https://github.com/Junjue-Wang/DisasterM3
Highlights
DisasterM3 includes 26,988 bi-temporal satellite images and 123k instruction pairs across… See the full description on the dataset page: https://huggingface.co/datasets/Patency/DisasterM3.DisasterM3
DisasterM3: A Remote Sensing Vision-Language Dataset for Disaster Damage Assessment and Response
Junjue Wang*,
Weihao Xuan*,
Heli Qi, Zhihao Liu, Kunyi Liu, Yuhan Wu, Hongruixuan Chen,
Jian Song
Junshi Xia, Zhuo Zheng, Naoto Yokoya†
* Equal Contributions
† Corresponding Author
Paper: https://arxiv.org/abs/2505.21089
Code: https://github.com/Junjue-Wang/DisasterM3
Highlights
DisasterM3 includes 26,988 bi-temporal satellite images and 123k instruction pairs across… See the full description on the dataset page: https://huggingface.co/datasets/Patty101/DisasterM3.DisasterM3
DisasterM3: A Remote Sensing Vision-Language Dataset for Disaster Damage Assessment and Response
Junjue Wang*,
Weihao Xuan*,
Heli Qi, Zhihao Liu, Kunyi Liu, Yuhan Wu, Hongruixuan Chen,
Jian Song
Junshi Xia, Zhuo Zheng, Naoto Yokoya†
* Equal Contributions
† Corresponding Author
Paper: https://arxiv.org/abs/2505.21089
Code: https://github.com/Junjue-Wang/DisasterM3
Highlights
DisasterM3 includes 26,988 bi-temporal satellite images and 123k instruction pairs across… See the full description on the dataset page: https://huggingface.co/datasets/hqaaaaaaa/DisasterM3.disaster-damage-assessment
Disaster Damage Assessment Dataset
Binary damage classification dataset (damage vs. no damage) across 4 natural disaster domains, used in:
Unsupervised Domain Adaptation for Rapid Disaster Damage Assessment
Dataset Structure
Domain
Code
Event
Images
Ecuador Earthquake
E
2016
1,724
Nepal Earthquake
N
2015
19,104
Hurricane Matthew
M
2016
333
Typhoon Ruby
R
2014
833
Total: 21,994 images across 4 domains.
Columns:
image: PIL Image
label: 0 =… See the full description on the dataset page: https://huggingface.co/datasets/abalhomaid/disaster-damage-assessment.DisasterM3
DisasterM3: A Remote Sensing Vision-Language Dataset for Disaster Damage Assessment and Response
Junjue Wang*,
Weihao Xuan*,
Heli Qi, Zhihao Liu, Kunyi Liu, Yuhan Wu, Hongruixuan Chen,
Jian Song
Junshi Xia, Zhuo Zheng, Naoto Yokoya†
* Equal Contributions
† Corresponding Author
Paper: https://arxiv.org/abs/2505.21089
Code: https://github.com/Junjue-Wang/DisasterM3
Highlights
DisasterM3 includes 26,988 bi-temporal satellite images and 123k instruction pairs across 5… See the full description on the dataset page: https://huggingface.co/datasets/CyberAltais/DisasterM3.Disaster_Recognition_RemoteSense_EN_CN_JA
xView2 Multi-Language Disaster Recognition Dataset
This dataset is derived from the xBD (xView2) Building Damage Assessment Dataset and has been reformatted for Vision-Language Model (VLM) training with multi-language support.
📊 Dataset Overview
This dataset contains satellite imagery paired with multi-language conversational annotations for disaster recognition tasks. It supports three languages: English, Chinese (中文), and Japanese (日本語).
Dataset Splits… See the full description on the dataset page: https://huggingface.co/datasets/xn67744/Disaster_Recognition_RemoteSense_EN_CN_JA.Crop-Flood-Disaster-Classification-Dataset
Crop Flood Disaster Classification Dataset
The current agricultural industry faces challenges of frequent flood disasters, making it difficult to quickly assess crop damage, affecting the stability of agricultural production and supply chains. Existing solutions largely rely on manual assessments which are inefficient and highly subjective, failing to meet the need for rapid response. This dataset aims to help AI models quickly assess damage by providing images of crops with… See the full description on the dataset page: https://huggingface.co/datasets/Harini1617/Crop-Flood-Disaster-Classification-Dataset.Disaster_Recognition_RemoteSense_EN_CN_JA
xView2 Multi-Language Disaster Recognition Dataset
This dataset is derived from the xBD (xView2) Building Damage Assessment Dataset and has been reformatted for Vision-Language Model (VLM) training with multi-language support.
📊 Dataset Overview
This dataset contains satellite imagery paired with multi-language conversational annotations for disaster recognition tasks. It supports three languages: English, Chinese (中文), and Japanese (日本語).
Dataset Splits… See the full description on the dataset page: https://huggingface.co/datasets/ly-ai/Disaster_Recognition_RemoteSense_EN_CN_JA.disaster36k_fold4
UAVDisaster36K Dataset
Overview
UAVDisaster36K is a large-scale aerial image dataset for disaster recognition collected from UAVs, drones, and other low-altitude aerial platforms. The dataset contains images of disaster and non-disaster scenarios across four categories:
Earthquake
Flood
Fire
Normal
The dataset was created to support research in aerial image classification, disaster monitoring, emergency response, and efficient deep learning models for UAV… See the full description on the dataset page: https://huggingface.co/datasets/ckyrkou/disaster36k_fold4.Disaster_Recognition_RemoteSense_EN_CN_JA
xView2 Multi-Language Disaster Recognition Dataset
This dataset is derived from the xBD (xView2) Building Damage Assessment Dataset and has been reformatted for Vision-Language Model (VLM) training with multi-language support.
📊 Dataset Overview
This dataset contains satellite imagery paired with multi-language conversational annotations for disaster recognition tasks. It supports three languages: English, Chinese (中文), and Japanese (日本語).
Dataset Splits… See the full description on the dataset page: https://huggingface.co/datasets/WayBob/Disaster_Recognition_RemoteSense_EN_CN_JA.DisasterM3
DisasterM3: A Remote Sensing Vision-Language Dataset for Disaster Damage Assessment and Response
Junjue Wang*,
Weihao Xuan*,
Heli Qi, Zhihao Liu, Kunyi Liu, Yuhan Wu, Hongruixuan Chen,
Jian Song
Junshi Xia, Zhuo Zheng, Naoto Yokoya†
* Equal Contributions
† Corresponding Author
Paper: https://arxiv.org/abs/2505.21089
Code: https://github.com/Junjue-Wang/DisasterM3
Highlights
DisasterM3 includes 26,988 bi-temporal satellite images and 123k instruction pairs across… See the full description on the dataset page: https://huggingface.co/datasets/Vane3RS/DisasterM3.DisasterM3
DisasterM3: A Remote Sensing Vision-Language Dataset for Disaster Damage Assessment and Response
Junjue Wang*,
Weihao Xuan*,
Heli Qi, Zhihao Liu, Kunyi Liu, Yuhan Wu, Hongruixuan Chen,
Jian Song
Junshi Xia, Zhuo Zheng, Naoto Yokoya†
* Equal Contributions
† Corresponding Author
Paper: https://arxiv.org/abs/2505.21089
Code: https://github.com/Junjue-Wang/DisasterM3
Highlights
DisasterM3 includes 26,988 bi-temporal satellite images and 123k instruction pairs across… See the full description on the dataset page: https://huggingface.co/datasets/Xunlun/DisasterM3.Crop-Flood-Disaster-Classification-Dataset
Crop Flood Disaster Classification Dataset
The current agricultural industry faces challenges of frequent flood disasters, making it difficult to quickly assess crop damage, affecting the stability of agricultural production and supply chains. Existing solutions largely rely on manual assessments which are inefficient and highly subjective, failing to meet the need for rapid response. This dataset aims to help AI models quickly assess damage by providing images of crops with varying… See the full description on the dataset page: https://huggingface.co/datasets/Mobiusi/Crop-Flood-Disaster-Classification-Dataset.DisasterM3
DisasterM3: A Remote Sensing Vision-Language Dataset for Disaster Damage Assessment and Response
Junjue Wang*,
Weihao Xuan*,
Heli Qi, Zhihao Liu, Kunyi Liu, Yuhan Wu, Hongruixuan Chen,
Jian Song
Junshi Xia, Zhuo Zheng, Naoto Yokoya†
* Equal Contributions
† Corresponding Author
Paper: https://arxiv.org/abs/2505.21089
Code: https://github.com/Junjue-Wang/DisasterM3
Highlights
DisasterM3 includes 26,988 bi-temporal satellite images and 123k instruction pairs across… See the full description on the dataset page: https://huggingface.co/datasets/Pandamx/DisasterM3.Disaster-Type_Classification_Dataset_for_Automated_Fact-Checking
DTCD-AFC: Disaster-Type Classification Dataset for Automated Fact-Checking
Overview
The DTCD-AFC is a dataset designed for disaster-type classification evaluation for automated fact-checking.
It consists of multimodal social media posts collected based on past natural disasters, each labeled with the disaster type to which its content relates.
The social media posts are sourced from CrisisMMD.
Files
disaster_type_classification_dataset_for_afc.csv: The CSV… See the full description on the dataset page: https://huggingface.co/datasets/o-yas/Disaster-Type_Classification_Dataset_for_Automated_Fact-Checking.DisasterM3
DisasterM3: A Remote Sensing Vision-Language Dataset for Disaster Damage Assessment and Response
Junjue Wang*,
Weihao Xuan*,
Heli Qi, Zhihao Liu, Kunyi Liu, Yuhan Wu, Hongruixuan Chen,
Jian Song
Junshi Xia, Zhuo Zheng, Naoto Yokoya†
* Equal Contributions
† Corresponding Author
Paper: https://arxiv.org/abs/2505.21089
Code: https://github.com/Junjue-Wang/DisasterM3
Highlights
DisasterM3 includes 26,988 bi-temporal satellite images and 123k instruction pairs across… See the full description on the dataset page: https://huggingface.co/datasets/yufaxia/DisasterM3.DisasterM3
DisasterM3: A Remote Sensing Vision-Language Dataset for Disaster Damage Assessment and Response
Junjue Wang*,
Weihao Xuan*,
Heli Qi, Zhihao Liu, Kunyi Liu, Yuhan Wu, Hongruixuan Chen,
Jian Song
Junshi Xia, Zhuo Zheng, Naoto Yokoya†
* Equal Contributions
† Corresponding Author
Paper: https://arxiv.org/abs/2505.21089
Code: https://github.com/Junjue-Wang/DisasterM3
Highlights
DisasterM3 includes 26,988 bi-temporal satellite images and 123k instruction pairs across… See the full description on the dataset page: https://huggingface.co/datasets/Ling200424/DisasterM3.
