Gait and Gesture Anonymization in Video Using Deep Learning

Develop a solution for manipulating the gait and/or gestures of people in videos, to preserve their privacy and protect them against person identification systems based on gait recognition

Required interest(s)

  • Video Recognition/Detection/Segmentation
  • Applied Deep Learning
  • Human-Centered and Community-Minded Information Systems

What do you get

  • A challenging assignment within a practical environment
  • Professional guidance
  • Courses aimed at your graduation period
  • Support from our academic Research center at your disposal

What you will do

  • 65% Research
  • 10% Analyze, design, realize
  • 25% Documentation

According to the General Data Protection Regulation (GDPR) approved by the European Commission, any information that can be attributed to an individual, or help identify them –on its own or in combination with other pieces of information– constitutes personal data.

One of the personal identifiers that falls under this definition, and that is generally overlooked, is the gait of a person: the combination of walking stride and cadence, or in other words, the way a person walks and moves. The gait, along with body gestures, could unequivocally identify a person, as it has been proven by the existence of gait-based person recognition systems.

Different methods have been proposed to tackle gait anonymization while preserving the naturalness of the human body motion, although no actual implementations have been made public. In addition, they still present some limitations, especially in terms of temporal smoothness, or the (high) quality that is generally required from the input video data to achieve a good performance.

A system capable of anonymizing gait, and even gestures, would also be valuable to protect personal identifiers like gender or age. Moreover, it could be especially powerful in combination with anonymization methods for other personal identifiers (such as faces), with the goal of creating a GDPR-compliant full anonymization system.

About Info Support Research Center

We anticipate on upcoming and future challenges and ensures our engineers develop cutting-edge solutions based on the latest scientific insights. Our research community proactively tackles emerging technologies. We do this in cooperation with renowned scientists, making sure that research teams are positioned and embedded throughout our organisation and our community, so that their insights are directly applied to our business. We truly believe in sharing knowledge, so we want to do this without any restrictions.

Sign up for this assignment

  • Geaccepteerde bestandstypen: docx, doc, txt, pdf.
  • Geaccepteerde bestandstypen: docx, doc, txt, pdf.

Other Mastertheses

graduation assignment

3D Background Reconstruction from 2D Videos Based on Biological Depth Cues

Develop a depth estimation method capable of learning human vision-based cues (such as semantic meaning, blurring, or texture), for its application on 2D background extraction and its reconstruction i…

graduation assignment

Transactional guarantees in a distributed system

There is a large body of research on ordering events and transactions in large distributed software systems. In the past, these large distributed systems were highly specialized and took special care …

graduation assignment

Gait and Gesture Anonymization in Video Using Deep Learning

Develop a solution for manipulating the gait and/or gestures of people in videos, to preserve their privacy and protect them against person identification systems based on gait recognition

graduation assignment

Scala code quality metrics (multi paradigm)

We have shown that we are able to use existing object oriented and functional metrics to predict bug density in Scala projects. This means there is a relation between different metrics and the possibi…