Cultural Heritage 3D Reconstruction for quality assessment (CH3D-Reco)

Dataset Description This dataset contains the data and resources required to reproduce and analyze the 3D model reconstruction and quality evaluation experiments presented in the article with DOI: https://doi.org/10.1016/j.image.2026.117514. The dataset includes: Original Model Captures: Synthetic images and masks generated with Blender from the original 3D models, used as input for reconstruction and evaluation. Blender Project Files: Scene files containing camera configurations, lighting setups, and rendering parameters necessary to replicate the image acquisition process. Reconstructed Models: 3D models obtained from the synthetic image sets using Structure from Motion (SfM) techniques. Camera and Capture Metadata: Detailed information about camera positions, orientations, and the image subsets used for reconstruction and quality assessment, enabling reproducibility. Evaluation Results: Spreadsheet files containing objective metric results, subjective evaluation scores, and interaction logs collected during the quality assessment experiments. This dataset is intended to support reproducibility, comparative evaluation, and further research in 3D reconstruction and quality assessment. Additional Resources Complementary materials related to the project are available at the GitHub repository:https://github.com/gti-upm/CH3D-Reco/ Further information can be found in the README.md file.