UTSA North Paseo Building


Vision & Artificial Intelligence Lab

Amanda Fernandez, Ph.D., Assistant Professor
Department of Computer Science, University of Texas at San Antonio
North Paseo Building 2.210


  • [8/2020] Joshua Greene has completed his Masters Project, "Message Passing Algorithms for Graphs as Input to Neural Networks" - congratulations!
  • [7/2020] Paper accepted to the OpenEDS Workshop at ECCV 2020: "EyeSeg: Fast and Efficient Few-Shot Semantic Segmentation" - congratulations Jon Perry!
  • [6/2020] VAIL welcomes Jenelle Millison for the UTSA CONNECT Summer Undergraduate Research Experience!
  • [5/2020] Soe Than has completed his Masters Project, "Deepfake Detection" - congratulations and best of luck at PwC AI!
  • [12/2019] Richard Tran has completed his Masters Thesis, "Defending and Detecting Against Adversarial Attacks" - congratulations!
  • [12/2019] KSAT interview on detecting deep fakes - watch here!
  • [9/2019] Poster accepted to WiML workshop at NeurIPS 2019!: "Visual Saliency Against Adversarial Examples"
  • [8/2019] Paper accepted to the Eye Tracking for VR and AR Workshop at ICCV 2019: "MinENet: A Dilated CNN for Semantic Segmentation of Eye Features" - congratulations Jon Perry!
  • [8/2019] Paper accepted to ISVC 2019: "On the Salience of Adversarial Examples"
  • [8/2019] Michael Geyer has completed his Masters Thesis, "Improved Feature Selection Using Neural Networks" - congratulations!
  • [8/2019] David Patrick has completed his Masters Thesis, "Generating Robust Data Sets and Models for Biological Motif Problems Using GANs" - congratulations!
  • [8/2019] VAIL is open! We are now located in NPB 2.210. Website under construction..


The UTSA Vision and Artificial Intelligence Lab (VAIL) focuses on basic research toward secure deep learning models.
VAIL constructs deep theoretical models - considering adversarial examples and cybersecurity approaches and applying these models to breadth of real-world technologies.

Secure Adversarial Learning

Adversarial networks as well as malicious input in the form of adversarial examples. Create models robust to evolving attacks.

Object Recognition & Segmentation

Attentional mechanisms for digital media, to recognize and locate salient data in complex real-world environments.

Deep Learning for the Physical Sciences

Reduce time and resource needs for complex experiments, such as fusion reaction in particle physics and applications of bioinformatics.

Our Lab


  • Joshua Greene - MS Project "Message Passing Algorithms for Graphs as Input to Neural Networks" - 2020
  • Richard Tran - MS Thesis "Defending and Detecting Against Adversarial Attacks" - 2020
  • Soe Than - MS Project "Deepfake Detection" - May 2020 [1st job: Machine Learning Engineer at PwC]
  • Tiffany Tsai - BS - May 2020 [1st job: USAA]

Interested in Collaborating?

Please review the following before reaching out.

UTSA students should be strong programmers with motivation and an interest in learning.
Previous coursework or experience in machine learning, data science, A.I., or related fields is a must.

  • UTSA students at the undergraduate, Masters, or Ph.D. levels are welcome to apply.
  • Minimum time committment is one year.
  • Independent Study or Directed Research for UTSA credit must be set up well in advance of the start of the semester.
  • At this time, there is no financial support for RAs. Any future opportunities will be advertised on this page.
Admissions to UTSA:
Please note that it is not possible for faculty to provide individual assessment of your chances of admission to our department. In particular, students are not admitted to the department by research project directors - therefore contacting individual faculty will not increase chances of admission. Due to the overabundance of such emails, I am unable to reply to these requests.

Find us on ...