University of Maryland Active Authentication-02 (UMDAA-02) Dataset

Phone Knows a lot + Illumination, pose and occlusion variation.

Overview

The UMDAA-02 data set consists of 141.14 GB of smartphone sensor signals collected from 48 volunteers on Nexus 5 phones over a period of 2 months (15 Oct. 2015 to 20 Dec. 2015). The data collection sensors include the front-facing camera, touchscreen, gyroscope, accelerometer, magnetometer, light sensor, GPS, Bluetooth, WiFi, proximity sensor, temperature sensor and pressure sensor. The data collection application also stored the timing of screen lock and unlock events, start and end time stamps of calls, currently running foreground application etc. The volunteers used the research phone as their primary device for a week and were given the option to stop data collection at will and review the stored data prior to sharing.

Downloads

Full UMDAA-02 Dataset:

The dataset is not available publicly anymore.

UMDAA-02 Face Dataset (UMDAA-02-FD):

A sampled version of the front-camera image captures of the UMDAA-02 dataset are annotated for face Detection and verification tasks. The annotated dataset can be downloaded from here.

Face detection and verification results on this data can be found in the following papers:
  1. U. Mahbub, S. Sarkar, and R. Chellappa. Partial face detection in the mobile domain. arXiv preprint arXiv:1704.02117, 2017.(pdf)
  2. U. Mahbub, S. Sarkar and R. Chellappa, "Pooling Facial Segments to Face: The Shallow and Deep Ends," 2017 12th IEEE International Conference on Automatic Face & Gesture Recognition (FG 2017), Washington, DC, 2017, pp. 634-641.(pdf)
  3. U. Mahbub, V. M. Patel, D. Chandra, B. Barbello and R. Chellappa, "Partial face detection for continuous authentication," 2016 IEEE International Conference on Image Processing (ICIP), Phoenix, AZ, 2016, pp. 2991-2995.(pdf)
  4. U. Mahbub, S. Sarkar, V. M. Patel and R. Chellappa, "Active user authentication for smartphones: A challenge data set and benchmark results," 2016 IEEE 8th International Conference on Biometrics Theory, Applications and Systems (BTAS), Niagara Falls, NY, 2016, pp. 1-8.(pdf)

UMDAA-02 Touch Dataset:

A sampled version of the touch-gesture data of the UMDAA-02 dataset are annotated for touch-based user recognition and verification tasks. The annotated dataset can be downloaded from here.
Benchmark results are presented in:
  1. U. Mahbub, S. Sarkar, V. M. Patel and R. Chellappa, "Active user authentication for smartphones: A challenge data set and benchmark results," 2016 IEEE 8th International Conference on Biometrics Theory, Applications and Systems (BTAS), Niagara Falls, NY, 2016, pp. 1-8.(pdf)

UMDAA-02 Location History Data:

Following paper/s contains bechmark results on location history-based user authentication on the UMDAA-02 dataset:
  1. U. Mahbub and R. Chellappa, "PATH: Person authentication using trace histories," 2016 IEEE 7th Annual Ubiquitous Computing, Electronics & Mobile Communication Conference (UEMCON), New York, NY, 2016, pp. 1-8.(pdf)

References

If you use this dataset, please cite our benchmark paper (pdf):
  • Plain Text:
    U. Mahbub, S. Sarkar, V. M. Patel and R. Chellappa, "Active user authentication for smartphones: A challenge data set and benchmark results," 2016 IEEE 8th International Conference on Biometrics Theory, Applications and Systems (BTAS), Niagara Falls, NY, 2016, pp. 1-8.
  • Bibtex:
    					@INPROCEEDINGS{Mahbub_Btas2016_UMDAA02,
					author={U. Mahbub and S. Sarkar and V. M. Patel and R. Chellappa},
					booktitle={2016 IEEE 8th International Conference on Biometrics Theory, Applications and Systems (BTAS)},
					title={Active user authentication for smartphones: A challenge data set and benchmark results},
					year={2016},
					pages={1-8},
					doi={10.1109/BTAS.2016.7791155},
					month={Sept},}

Number of Visitors

Change History

  1. 30 May, 2018: Website framework created.
  2. 31 May, 2018: Downloadable linkes added.
  3. 09 July, 2018: Added visitor counter.
  4. 06 June, 2019: Dataset request emails added.



Last Modified: 31 May, 2018. Please direct comments to Upal Mahbub