Skip to main content

Evaluation tools for verification systems under spoofing attacks: examples in face verification

Project description

This package provides methods for evaluation of biometric verification systems under spoofing attacks. The evaluation is based on the Expected Performance and Spoofability Curve (EPSC). Using this package, you can compute thresholds based on EPSC, compute various error rates and plot various curves related to EPSC.

Besides providing methods for plotting EPSC within your own scripts, this package brings several scripts that you can use to evaluate your own verification system (fused with an anti-spoofing system or not) from several perspectives. For example, you can:
  • evaluate the threshold of a classification system on the development set

  • apply the threshold on an evaluation or any other set to compute different error rates

  • plot score distributions

  • plot different performance curves (DET, EPC and EPSC)

Furthermore, you can generate hypothetical data and use them to exemplify the above mentioned functionalities.

If you use this package and/or its results, please cite the following publication:

  1. Our original paper on biometric evaluation (title, pdf and bibtex to be announced soon):

    @ARTICLE{Chingovska_IEEETIFS_2014,
       author = {Chingovska, Ivana and Anjos, Andr{\'{e}} and Marcel, S{\'{e}}bastien},
       title = {Biometrics Evaluation Under Spoofing Attacks},
       journal = {IEEE Transactions on Information Forensics and Security},
       year = {2014},
    }
  2. Bob as the core framework used to run the experiments:

    @inproceedings{Anjos_ACMMM_2012,
        author = {A. Anjos AND L. El Shafey AND R. Wallace AND M. G\"unther AND C. McCool AND S. Marcel},
        title = {Bob: a free signal processing and machine learning toolbox for researchers},
        year = {2012},
        month = oct,
        booktitle = {20th ACM Conference on Multimedia Systems (ACMMM), Nara, Japan},
        publisher = {ACM Press},
    }

If you wish to report problems or improvements concerning this code, please contact the authors of the above mentioned papers.

Installation

Installation

The antispoofing.evaluation package is a satellite package of the free signal processing and machine learning library Bob. This dependency has to be downloaded manually. This version of the package depends on Bob version 2 or greater. To install packages of Bob, please read the Installation Instructions. For Bob to be able to work properly, some dependent Bob packages are required to be installed. Please make sure that you have read the Dependencies for your operating system.

The most simple solution is to download and extract antispoofing.evaluation package, then to go to the console and write:

$ cd antispoofing.evaluation
$ python bootstrap-buildout.py
$ bin/buildout

This will download all required dependent Bob and other packages and install them locally.

Using the package

After instalation of the package, go to the console and type:

$ ./bin/sphinx-build docs sphinx

Now, the full documentation of the package, including a User Guide, will be availabe in sphinx/index.html.

Problems

In case of problems, please contact ivana.chingovska@idiap.ch

Project details


Download files

Download the file for your platform. If you're not sure which to choose, learn more about installing packages.

Source Distribution

antispoofing.evaluation-2.0.0.zip (56.1 kB view details)

Uploaded Source

File details

Details for the file antispoofing.evaluation-2.0.0.zip.

File metadata

File hashes

Hashes for antispoofing.evaluation-2.0.0.zip
Algorithm Hash digest
SHA256 29b8246ab02bfd706780d2750da22402da6eb8dfea5ae800b1671267239ada3d
MD5 ff73c02ecb6b43a4369826a7263b3211
BLAKE2b-256 4ac42ea60446b5dba1acf10cf37d51e199311ae362db5edc08235df15aacf195

See more details on using hashes here.

Supported by

AWS AWS Cloud computing and Security Sponsor Datadog Datadog Monitoring Fastly Fastly CDN Google Google Download Analytics Microsoft Microsoft PSF Sponsor Pingdom Pingdom Monitoring Sentry Sentry Error logging StatusPage StatusPage Status page