The Geo-metric dataset is a part of a paper titled: Geo-metric: A Perceptual Dataset of Distortions on Faces. Wolski, Trutoiu, Dong, Shen, MacKenzie, Chapiro ACM Transactions on Graphics, Proceedings of SIGGRAPH Asia 2022
If you use this data, please cite the paper mentioned above in your publication. Dataset consists of five primary components:
- Directory containing stimuli used in the subjective study described in the paper (stimuli).
- Directory containing low resolution meshes used for stimuli generation (lowpoly).
- Directory containing matlab scripts that allow JOD scaling of the attached results and computing correlation coefficients for metrics mentioned in the paper (analysis).
- CSV file providing general information on the meshes (Dataset.csv).
- CSV file containing responses from the conducted subjective study (SubjectiveData.csv)
Please see our instructions.pdf file for additional details.
Please feel free to reach out to Krzysztof Wolski (wolskikrzys@gmail.com) and Alex Chapiro (alex@chapiro.net) with any questions or comments, we love hearing about folk using our data!
This work is licensed under a Creative Commons Attribution-NonCommercial-ShareAlike 4.0 International License.