datasets
Training and evaluation data, with the modality, task and licence stated up front. Listed live from the Hugging Face Hub.
forest-fire-dataset
Forest Fire Dataset
Dataset Description
This dataset contains frames extracted from forest fire videos for computer vision and machine learning applications. The dataset is designed for tasks such as fire detection, smoke detection, and wildfire monitoring.
Dataset Structure
The dataset is organized into three splits following standard machine learning practices:
forest-fire-dataset/
├── train/
│ ├── train_00000-08999/
│ ├── train_09000-17999/
│… See the full description on the dataset page: https://huggingface.co/datasets/Bruhtian/forest-fire-dataset.ForestFireFIghting_vision
ForestFireFighting-Vision
A vision dataset for aerial forest-fire risk assessment. Each sample pairs a top-down
rendered forest-fire scene with a machine-generated, structured risk analysis: the image is
decomposed into coarse rectangular risk zones, each carrying a risk level, the fire
points it contains, and a natural-language justification in Chinese.
The dataset targets models that must go beyond detection — i.e. reproduce an operator-style
reasoning trace: where the risk is… See the full description on the dataset page: https://huggingface.co/datasets/DONKEY13479/ForestFireFIghting_vision.forest-fire-annotations
Forest Fire Detection Dataset — Auto-Annotated
Bounding-box annotated version of touati-kamel/forest-fire-dataset,
built for training forest-fire / smoke / fog object detection models.
Overview
This dataset contains video frames auto-labeled with bounding boxes for fire and
smoke-related visual phenomena, using a zero-shot open-vocabulary object detector
(Grounding DINO). It is derived from the original touati-kamel/forest-fire-dataset image
classification dataset… See the full description on the dataset page: https://huggingface.co/datasets/touati-kamel/forest-fire-annotations.forest_fireThe initial pictures from UAV (FLAME dataset, file 9) with the size 3840 × 2160 were splitted in nine non-overlapping parts, each such part having a size of 1280 × 720 pixels. Information about locations with fires was extracted from masks, prepared by developers of primary dataset for segmentation (FLAME, file 10), then, with the help of marking algorithm of connected components the data about points with fire were extracted: information about their sizes and coordinates are getting saved in… See the full description on the dataset page: https://huggingface.co/datasets/alarmod/forest_fire.ForestFireFIghting_plan
Forest Fire Fighting UAV Mission Planning
An instruction-tuning dataset for multi-UAV forest-fire fighting mission planning. Each
example gives a model a complete situational prompt — the drone command API, the available
UAV fleet, and the zone/fire-point layout of the fire scene — and asks it to emit an
executable Python mission chain that dispatches the fleet: extinguishing drones bomb
every fire point and then sweep their zones, surveillance drones cover the remaining zones… See the full description on the dataset page: https://huggingface.co/datasets/DONKEY13479/ForestFireFIghting_plan.forest-fire-datasetYOLO_ForestfireForestFireInsights-Eval
Dataset Card: ForestFireInsights-Eval
Model Description
This dataset contains data from the Center for Wildfire Research at University of Split (Website, Paper)
and original UAV images from the student team Evolonic at FAU Erlangen and Fraunhofer IISB.
We thank Lara Schindhelm, Sebastian Wiederhold and Adrian Sauer for gathering images with various drones! The dataset is formatted for use with our ForestFireVLMs (7B and 3B)
Citations
This model is… See the full description on the dataset page: https://huggingface.co/datasets/leon-se/ForestFireInsights-Eval.TNFD-ForestFireAlerts-21-24forest_fire_dataset-llama3forest-fire-dataset