The SeracFallDet dataset is a specialized sub-task of change detection focused on identifying volumetric changes. Specifically, this dataset contains images with serac fall (glacier ice collapse) annotations.
It consists of annotated pairs of high resolution images captured using fixed terrestrial time-lapse cameras. Detecting serac falls is critical for natural hazard monitoring, enabling better prevention of catastrophic events. Each annotation are masks generated by bounding boxes fusion.
Composition.
The dataset includes 11 scenes from glaciers (currently only within the French Alps), comprising:
- 8199 pre-registered images
- 2123 annotations of serac falls, represented as polygonal annotations (derived from bounding boxes) around detected changes, if any.
Acknowledgements.
We extend our sincere gratitude to all following contributors for providing access to the glacier footage that forms the dataset: Voltalia, DDT-74 and IGE, communauté de commune de la Vallée de Chamonix Mont-Blanc (CCVCMB) and COMPAGNIE DU MONT BLANC/Aiguille du Midi www.montblancnaturalresort.com).
The SeracFallDet dataset is distributed as compressed .zip archives for each split, organized hierarchically as follows:
SeracFallDet
├── train
│ ├── scene_1
│ │ ├── year_1
│ │ │ ├── images
│ │ │ └── label
│ │ ├── ...
│ │ └ position.json
│ └── ...
├── val
│ ├── scene_x
│ │ ├── year_x
│ │ │ ├── images
│ │ │ └── label
│ │ ├── ...
│ │ └ position.json
│ └── ...
└── test
├── scene_y
│ ├── year_y
│ │ ├── images
│ │ └── label
│ ├── ...
│ └ position.json
└── ...
Each label file is in .json format, containing the keys:
- filepath_1 (image name, pre-change)
- filepath_2 (image name, post-change)
- absolute_box (detected change)
The bounding boxes are provided in xyxy format, and masks can be generated online using these coordinates.
Coarse camera positions are stored in the position.json files.