“Signal Processing in the AI era” was the tagline of this year’s IEEE International Conference on Acoustics, Speech and Signal Processing, taking place in Rhodes, Greece.
In this context, Brent de Weerdt, Xiangyu Yang, Boris Joukovsky, Alex Stergiou and Nikos Deligiannis presented ETRO’s research during poster sessions and oral presentations, with novel ways to process and understand graph, video, and audio data. Nikos Deligiannis chaired a session on Graph Deep Learning, attended the IEEE T-IP Editorial Board Meeting, and had the opportunity to meet with collaborators from the VUB-Duke-Ugent-UCL joint lab.
Featured articles:
- De Weerdt, B. et al., “Designing Transformer Networks for Sparse Recovery of Sequential Data Using Deep Unfolding” (https://ieeexplore.ieee.org/abstract/document/10094712)
- Yang X. et al., “Relevance Propagation through Deep Conditional Random Fields” (https://ieeexplore.ieee.org/abstract/document/10095075)
- Stergiou A. et al., “Play it back: Iterative attention for audio recognition” (https://ieeexplore.ieee.org/abstract/document/10096532, developed at the University of Bristol)
- First PhD in collaboration with Technical University Twente about ISFETs. (Luc Bousse).
- Oscar Steenhaut – Rector at VUB.
We are very proud of Brent De Weerdt (prom N. Deligiannis), Joris Wuts (prom. J. Vandemeulebroucke) and Silvia Zaccadi (prom. B. Jansen) who received their scholarship from FWO aspirant strategic basic research for the next 2+2 years. Way to go!
On October 6th 2025 at 16:00, Ran Zhao will defend their PhD entitled “DEEP LEARNING-BASED HUMAN POSTURE NORMALIZATION AND AUTOMATIC ANTHROPOMETRIC MEASUREMENT”.
Everybody is invited to attend the presentation in room D.2.01 or online via this link.
Accurate and user-friendly anthropometric measurement remains a major challenge in computer vision, as existing approaches typically require controlled scanning conditions, standard postures, or unclothed bodies. These constraints limit their usability in practical scenarios.
This thesis proposes a sequence of deep learning-based solutions to overcome these limitations. We first introduce OrienNormNet, an iterative network for robust orientation normalization, ensuring that scans are consistently aligned without manual preprocessing. Building on this, PoseNormNet is presented as the first posture normalization framework that transforms arbitrarily posed scans into a canonical T-pose while preserving identity details, removing the need for skeleton rigging. Next, W2H-Net demonstrates the feasibility of directly estimating the waist-to-hip ratio from partial dressed scans, showing that reliable indicators can be derived even from incomplete data. Finally, MeasureXpert provides a breakthrough toward real-world usability: it enables automatic extraction of anthropometric measurements from only two unregistered, partial, and clothed scans acquired in arbitrary poses.
To support these developments, the BWM dataset was synthesized for training, validation, and evaluation. Comprehensive experiments on both synthetic and real-world data confirm the effectiveness and robustness of the proposed methods. Collectively, the contributions progressively address key challenges related to cost, posture, and clothing, moving the field closer to practical, flexible, and accessible body measurement solutions.
The algorithms presented in this thesis have been disseminated through prestigious journals and conferences, demonstrating a modest yet meaningful impact on both academic research and industrial applications.
On February 5 2021 at 16.00 Alexander Soñora Mengana will defend his PhD entitled “COMPUTER AIDED DETECTION OF LUNG NODULES FROM CT IMAGING”.
Lung cancer is the first cause of cancer related death worldwide Early detection can have substantial impact on treatment outcome Computer aided detection (systems can play an important role in improving the detection rate and reducing the clinical workload, in particular considering the lung cancer screening protocols that are currently being set up.
Starting from an existing system for computer aided detection of lung cancer, several aspects of the processing pipeline were investigated, with the aim to improve the accuracy and robustness of the process The system employed a two stage approach, comprising of a candidate detector, and a false positive reduction step based on hand crafted features Initially the design of the system was changed to a modular architecture to facilitate introducing alterations at different stages, and evaluate their impact The system’s efficiency and usability was improved and individual components were tuned.
Next, a thorough characterisation of its performance by participating to the LUNA 16 Challenge The participation implied training and testing the system on large clinical dataset It also enabled the objective comparison to other proposed CAD approaches using a common evaluation methodology The system as a whole, was shown to perform well, achieving comparable results to other full system submissions at the time of the challenge Closer analysis, revealed this was mainly due to a sensitive nodule candidate detector, whereas other approaches were found to have better false positive reduction.
Subsequently, several aspects of the pipeline were investigated, to improve on this baseline results An improved lung segmentation procedure was added to the preprocessing stage The method reduces the amount of failed lung segmentations due to artefacts or even tracheotomy by performing an error detection and correction procedure, making the candidate detection process more robust.
Candidate detectors often mark a large number of non nodule structures compared to the of actual nodules This imbalance in the data during training, may hinder the performance of the classifier Data balancing methods, comprising both undersampling and oversampling approaches in feature space, were therefore investigated in detail Surprisingly, undersampling the majority class, as performed in the original system, was found to perform worse compared to no balancing Balancing by oversampling the minority class allowed to improve the result further
Over the course of my PhD, deep learning methods emerged for medical image analysis, and rapidly outperformed alternative approaches in the CAD domain I therefore investigated how to increase the accuracy of the false positive reduction by training a convolutional neural network using the candidates provided by my detector, and obtained a substantial increase in accuracy
Interestingly, combining the learnt features with the hand crafted features, improved the results even further
The Master in Applied Computer Science at VUB accepts students with a bachelor or Master in Engineering or Exact Sciences
- May 1968: student revolt impacting citizen’s mindset, and changing society as well as university governance.
- 1968-1969 The unitary ULB/VUB starts a curriculum in mechanical/electrotechnical engineering in Dutch.
- 1969-1970: Prof. (Major) Marcel Verlinden teaches Electronics on the ULB-Campus Solbosch. While vacuum tubes and sometimes bipolar transistors are still the predominant devices, he teaches MOSFETs.
- 1969-1970 VUB – ULB split: VUB becomes an independent university – in the aftermath of “Leuven Vlaams”, Belgium gradually becomes a federal state between 1970 (3 communities are created) and 2014.
- 1970-1971: Oscar Steenhaut joins VUB from Louisiana State University – Baton Rouge, and Jacques Tiberghien becomes the first full time assistant.