mpg-ranch/light-stable-semantics

Light Stable Semantics Dataset

Dataset Description

This dataset contains aerial orthomosaic tiles captured at three different times of day (10:00, 12:00, and 15:00). Each tile includes the original RGB images, a co-registered canopy height model (CHM) raster, and pre-computed DINOv3 embeddings extracted using facebook/dinov3-vitl16-pretrain-sat493m. The dataset is designed for adapting vision encoders that can maintain consistent feature representations despite changes in illumination, with applications in remote sensing and environmental monitoring.

Dataset Features

Each record in the dataset contains the following features:

Feature	Type	Shape	Description
idx	string	-	Tile identifier in format {ROW}_{COL} for geographic referencing
image_t0	Image	1024×1024×3	Morning capture at 10:00 AM (time=1000)
image_t1	Image	1024×1024×3	Noon capture at 12:00 PM (time=1200)
image_t2	Image	1024×1024×3	Afternoon capture at 3:00 PM (time=1500)
cls_t0	float32	[1024]	DINOv3 CLS token (global features) for morning image
cls_t1	float32	[1024]	DINOv3 CLS token (global features) for noon image
cls_t2	float32	[1024]	DINOv3 CLS token (global features) for afternoon image
patch_t0	float32	[196, 1024]	DINOv3 patch tokens (spatial features) for morning image
patch_t1	float32	[196, 1024]	DINOv3 patch tokens (spatial features) for noon image
patch_t2	float32	[196, 1024]	DINOv3 patch tokens (spatial features) for afternoon image
canopy_height	int32	[1024, 1024]	Canopy height grid in centimetres derived from the canopy height model

The canopy height layer is reprojected to align with the RGB tiles and multiplied by 100 before casting to int32, so each value represents centimetres above ground. Missing data is encoded with -2147483648 (the minimum 32-bit integer).

The dataset is partitioned with an 80%/20% train/test split.

Usage Example

from datasets import load_dataset

# Load the dataset
dataset = load_dataset("mpg-ranch/light-stable-semantics")

# Access a single training record
sample = dataset['train'][0]

# Images for the three time points
morning_image = sample['image_t0']
noon_image = sample['image_t1'] 
afternoon_image = sample['image_t2']

# Pre-computed DINOv3 embeddings
morning_cls = sample['cls_t0']     # Global features (1024-dim)
noon_cls = sample['cls_t1']        # Global features (1024-dim)
afternoon_cls = sample['cls_t2']   # Global features (1024-dim)

morning_patches = sample['patch_t0']  # Spatial features (196×1024)
noon_patches = sample['patch_t1']     # Spatial features (196×1024)
afternoon_patches = sample['patch_t2'] # Spatial features (196×1024)

# Tile location identifier
tile_id = sample['idx']  # Format: "{ROW}_{COL} of tiles within the original orthomosaic"

# Co-registered canopy height (centimetres stored as int32)
canopy_cm = sample['canopy_height']

# Held-out evaluation tile
test_sample = dataset['test'][0]

Pre-computed Embeddings

The dataset includes pre-computed embeddings extracted using the facebook/dinov3-vitl16-pretrain-sat493m model:

• CLS Tokens: 1024-dimensional global feature vectors that capture scene-level semantics
• Patch Tokens: 196×1024 arrays encoding spatial relationships and local features
• Purpose: Enable efficient training and analysis without requiring on-the-fly feature extraction
• Model Details: DINOv3 Vision Transformer Large (16×16 patches) pre-trained on satellite imagery

Dataset Information

• Location: Lower Partridge Alley, MPG Ranch, Montana, USA
• Survey Date: November 7, 2024
• Coverage: 620 complete tile sets (80% train / 20% test split via seeded random sampling)
• Resolution: 1024×1024 pixels at 1.2cm ground resolution
• Total Size: ~6.4GB of image data plus embeddings
• Quality Control: Tiles with transient objects, such as vehicles, were excluded from the dataset. RGB imagery and canopy rasters are removed together to keep modalities aligned.

Use Cases

This dataset is intended for:

• Developing vision encoders robust to lighting variations
• Semantic stability research in computer vision
• Time-invariant feature learning
• Remote sensing applications requiring lighting robustness
• Comparative analysis of illumination effects on vision model features

Citation

If you use this dataset in your research, please cite:

@dataset{mpg_ranch_light_stable_semantics_2024,
  title={Light Stable Semantics Dataset},
  author={Kyle Doherty and Erik Samose and Max Gurinas and Brandon Trabucco and Ruslan Salakhutdinov},
  year={2024},
  month={November},
  url={https://huggingface.co/datasets/mpg-ranch/light-stable-semantics},
  publisher={Hugging Face},
  note={Aerial orthomosaic tiles with DINOv3 embeddings for light-stable semantic vision encoder training},
  location={MPG Ranch, Montana, USA},
  survey_date={2024-11-07},
  organization={MPG Ranch}
}

License

This dataset is released under the Creative Commons Attribution 4.0 International (CC BY 4.0) license.

Attribution Requirements:

• You must give appropriate credit to MPG Ranch
• Provide a link to the license
• Indicate if changes were made to the dataset