Moving Masks Presentation Attack (MMPA) Database

We introduce the Moving Mask Presentation Attack (MMPA) database, a face biometrics database collected using five different cameras. The data acquisition takes place in an indoor corridor setting, featuring both natural sunlight and artificial lighting to simulate realistic indoor environments and border control scenarios. The data subjects are captured on the move, walking towards the camera from a distance of six meters. Figure 1 provides an example illustrating the “on-the-move” capture setup. The camera is mounted on a stand, and participants are instructed to approach it from six meters away. The MMPA dataset consists of 15 data subjects, including 8 males and 7 females. 

Bona fide data capture 

Bona fide data subjects are instructed to walk towards the camera at a normal walking speed, while data is captured continuously. Each walk typically lasts between 40 to 70 seconds. Data collection is conducted across multiple sessions over a period of 7 days. In each session, 5 to 10 video samples are recorded for each subject using all five cameras in two different settings (a) with eye glasses (b) without eye glasses. Table 1 provides an overview of the bona fide data statistics collected across all five cameras. Table 2 shows the specifications of the cameras employed in this work. Figure 1 provides an example illustrating the “on-the-move” capture setup. The camera is mounted on a stand, and participants are instructed to approach it from six meters away. The MMPA dataset consists of 15 data subjects, including 8 males and 7 females. 

Presentation Attack data capture 

In this work, we focus on custom silicone face masks that align with real-world operational scenarios. To enhance the realism and effectiveness of presentation attacks, we introduce accessories such as eyeglasses, hair wigs, caps, and hoodies, which help make the attacks more convincing and difficult to detect. We employ four custom silicone face masks, previously shown to exhibit a high attack potential against COTS systems. 

The presentation attack samples are captured in four different cases, each designed to increase realism and challenge detection systems: Case 1: The attacker wears a silicone mask with a head cap and a hair wig, fully covering the mask. Case 2: The attacker wears a silicone mask in conjunction with eyeglasses, a head cap and a hair wig, further concealing the facial mask for a more realistic attack. Case 3: The attacker wears a silicone mask with eyeglasses, a head cap, and a hoodie, covering the facial mask to increase disguise complexity. Case 4: The attacker wears a silicone face mask with a hoodie and head cap, ensuring that the mask discontinuities remain hidden. These four cases are designed to create highly realistic and challenging presentation attacks, making them difficult to detect even for human observers, such as border guards, and monitoring on-the move biometric corridors. 

Figure 2 illustrates examples of bona fide and presentation attacks across all four cases with Device 5 

Copyright of MMPA :

The researcher can access the Moving Mask Presentation Attack (MMPA) database by following the procedure mentioned: 

• Researchers are required to send the request to use MMPA by sending a copy of the license agreement, completely filled, to vetrekarnarayan@unigoa.ac.in with the subject line "License agreement for Moving Mask Presentation Attack (MMPA) Database". 

Note: The license agreement has to be signed by the researcher or supervisor and the signature of a legal authority on behalf of the institution, such as the Head of the institution or Registrar, along with the institutional seal (Rubber Stamp).  The license agreement should be on the researchers institutional letterhead. 

The request will not be considered if due procedures are not followed.  

• Every request to avail MMPA will be placed before the Ethical Committee of the Institute for approval, and the confirmation of the request will be sent via email. Further, all the instructions related to access to the database will be detailed in the email.

This database is available only for research and educational purposes and not for any commercial use. Further, the use of the database is only for one year from the date of signing the agreement by the researcher. Beyond the allocated one year for the usage of data, the applicant has to apply again following the due procedure.  

The database is available only for research and educational purposes. All the rights of the Moving Mask Presentation Attack (MMPA) database are reserved, and commercial use/distribution of this database is strictly prohibited. All the technical reports and papers that report experimental results from this database should provide due acknowledgement and reference. If you use the database in any publications or reports, you must refer to the following paper:  

• Raghavendra Ramachandra, Narayan Vetrekar, Krishna Patel, Marissa Ataide, Sushma Venkatesh, and Rajendra Gad, Moving Masks: A Preliminary Study on Face Presentation Attack On-The-Move, WACV 2026 – Workshop on Manipulation, Generative, Adversarial, and Presentation Attacks in Biometrics, Tucson, Arizona.(Accepted).