Computer Science

Faculty of Engineering, LTH


Additional CS MSc Thesis Zoom Presentations 21 June 2021


From: 2021-06-21 13:15 to 14:00
Place: Online via zoom
Contact: birger [dot] swahn [at] cs [dot] lth [dot] se
Save event to your calendar

Two more Computer Science MSc theses to be presented on 21 June via Zoom

Monday, 21 June there will be two more master thesis presentations in Computer Science at Lund University, Faculty of Engineering.

The presentations will take place via Zoom, see link under the presentation.

Note to potential opponents: Register as an opponent to the presentation of your choice by sending an email to the examiner for that presentation ( Do not forget to specify the presentation you register for! Note that the number of opponents may be limited (often to two), so you might be forced to choose another presentation if you register too late. Registrations are individual, just as the oppositions are! More instructions are found on this page.


Presenters: Josef Abdul Al, Daniel Siljegovic Persson
Title: Evaluation of an RGB-D data based Gaze Tracker for Human-Robot Interaction
Examiner: Jacek Malec
Supervisor: Elin Anna Topp (LTH)

Eye gaze and body language is a very important part in interaction between people. The eye gaze especially is a clear indicator of where a person's attention is directed. For a robot to be able to communicate effectively with humans it is important to find the person's gaze direction. The purpose of this master thesis is to implement and evaluate a gaze tracker for use in human-robot interaction. This is done by using a pipeline of convolutional neural networks to extract facial features from an RGB-D image, and using those features to estimate head pose and gaze direction.

Link to presentation:

Link to popular science summary: To be updated


Presenters: Karl-Oskar Rikås, Frank Weslien
Title: Filtering False Positive Alarms in JavaDL and Language Experience Report
Examiner: Niklas Fors
Supervisor: Christoph Reichenbach (LTH), Alexandru Dura (LTH)

JavaDL is a domain-specific language (DSL) for writing static program analyses in declarative logic programming, based on Datalog. The key feature of this DSL is the ability to pattern-match on literal source code syntax and reason non-locally through declarative programming. Static program analyses generally suffer from producing false positive alarms. This results in developers having to deal with unnecessary alarms. A machine learning model could mitigate this problem by filtering false positive alarms. We investigate if features based on JavaDL's pattern-matching are effective.. Our results show that they are not, as the knowledge learned does not transfer over to unseen projects. Points-to analysis is another way of improving the precision of otherwise more conservative analysis. As the first users of JavaDL we wrote a Points-to analysis, for a subset of the Java language. We report on our experience and put forth possible improvements to JavaDL in a case study.

Link to presentation:

Link to popular science summary: To be updated

How to get to computer science