Images and Point Clouds Cultural Heritage Dataset: Photogrammetric Images and 3D Point Clouds of Five Historic Buildings with Semantic Labels (HEIG-VD mirror)

Provider
Eugenio Pellis, Andrea Masiero, Michele Betti, Grazia Tucci, Pierre Grussenmeyer (original authors); HEIG-VD (mirror)
License
CC-BY-4.0 (source)
Size
6,510 files
Format
PNG (in sampled files)
Stored at
Hugging Face (https://huggingface.co/datasets/heig-vd-geo/ImagesAndPointCloudsCulturalHeritageDataset)
Access
Free and public. No account with the data provider is needed.
DOI
10.5281/zenodo.14626611
Category
Earth Observation & Geospatial
Source
https://huggingface.co/datasets/heig-vd-geo/ImagesAndPointCloudsCulturalHeritageDataset

About this dataset

Photogrammetric images and 3D point clouds of five cultural heritage buildings, with semantic ground truth annotations for both the 2D images and the 3D points following the ARCHdataset architectural taxonomy (10 classes). The original release has five scene folders of about 35.5 GB with calibrated images, camera intrinsic and extrinsic parameters, point clouds and 2D and 3D labels, supporting 2D and 3D semantic segmentation, image to point cloud label transfer and multimodal fusion benchmarks for heritage building documentation and HBIM. This Hugging Face copy, kept by the HEIG-VD geomatics group, mirrors the five scene folders (1_SC, 2_OSA, 3_SS, 4_CG, 5_CB). The authoritative source is the Zenodo record by Eugenio Pellis, Andrea Masiero, Michele Betti, Grazia Tucci and Pierre Grussenmeyer, DOI 10.5281/zenodo.14626611, licensed CC BY 4.0. Cite the dataset and its paper.

How to load this dataset in Python

from huggingface_hub import hf_hub_download
path = hf_hub_download(repo_id="heig-vd-geo/ImagesAndPointCloudsCulturalHeritageDataset", filename="1_SC/1_Images/1_RGB/1_SC_001.png", repo_type="dataset")
print(path)

Data structure

File types in sampled files: PNG

Files

Example file: 1_SC/1_Images/1_RGB/1_SC_001.png

Analyze this dataset without downloading it

On Scigantic, Images and Point Clouds Cultural Heritage Dataset: Photogrammetric Images and 3D Point Clouds of Five Historic Buildings with Semantic Labels (HEIG-VD mirror) opens in a hosted JupyterLab notebook with the data already mounted, so you can read files where they are stored instead of copying them first. Each notebook includes an AI assistant that can see the mounted files and write and run analysis code. Open the dataset page to launch a notebook, or browse the full catalog.

This page is also available as Markdown and as JSON.

Related Earth Observation & Geospatial datasets