Face2Cam is a webcam user distance estimation dataset meant to help address user-to-camera distance estimation. The dataset was collected in lab with 50 participants. We used a CNN model to test the dataset performance. The code for model training as well as the dataset itself are available below.
The dataset contains 90 full-face images for each participant. Images were taken at six different distances from the user (10cm, 20cm, 30cm, 40cm, 50cm, 60cm), five yaw angles from face center (-90 degrees, -45 degrees, 0 degrees. 45 degrees, 90 degrees ), and three altitude angles (-15 degrees, 0 degrees, 15 degrees). Images are stored in a directory and named using a structured format: pXXX-(angle)-(distance)cm-(altitude).jpg
The code we used for data processing and model testing is available on GitHub
Data Processing CodeWe are happy to share the data with anyone interested! Please fill out the form below and we will send you the download link. Please make sure to use your institutional email. If you have any questions, please reach us at robbinslab@richmond.edu.
Download DatasetPlease use the following citation if you are using the dataset. Thank you!
@inproceedings{zhai2026Face2Cam,
author = {Zhai, Yixuan and Robbins, Arryn},
title = {Face2Cam: A User-to-Webcam Distance Estimation Image Dataset},
year = {2026},
isbn = {TBA},
publisher = {SCIEPRESS},
address = {Setubal, Portugal},
url = {TBA},
doi = {TBA},
booktitle = {Proceedings of the 21st International Conference on Computer Vision Theory and Application},
articleno = {325},
numpages = {8},
keywords = {Webcam-based interaction, Distance estimation, User-to-camera distance, Machine learning},
location = {Marbella, Spain},
series = {VISAPP '26}
}
We are happy to answer any of your questions! Please send an email to robbinslab@richmond.edu, and we will do our best to address your questions.