“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.
On July 2 2021 at 16.30 Abel Diaz Berenguer will defend his PhD entitled “Learning to predict human behavior in crowded scenes”.
Automatically understanding human behavior is one of the most fundamental research topic towards socially aware vision-based autonomous systems. There is an increasing interest in incorporating the social signal perspective into the learning systems pipeline. This dissertation focuses on developing and incorporating computational mechanisms of Computer Vision and Machine Learning to analyze and predict human behavior in crowded scenes automatically. Our research specifically addresses public safety assisted by autonomous video surveillance systems aiming to decrease the human labor dedicated to video monitoring.
Our research efforts concentrate on the information processing pipeline for learning systems that cope with human trajectory prediction and human behavior analysis in crowded scenes. We contribute to human trajectory prediction in crowded scenes with
(i) a novel latent variable model aware of the human-human and human-contextual interactions to predict plausible trajectories. Furthermore,
(ii) a novel latent location-velocity recurrent model that predicts future variable and feasible trajectories. Towards human anomalous behavior detection, we adopt two unsupervised approaches based on the scene dominant behavior and trajectories underlying properties to address trajectory-based anomaly detection. Besides, we contribute with
(iii) a supervised approach capable of attaining discriminative sequence-based feature representations to recognize whether video sequences depict violent human behavior.
Extensive experiments on publicly available datasets, demonstrate the effectiveness of our proposals.
Sofia Granda attended the Master in Biomedical Engineering in 2020-2021. She chose the program because she really liked mathematics, physics, and biology at high school and liked to be able to find practical solutions to problems. Sofia described the program in the following three words: Empathy, Logic and Medicine. Strengths of the program were the flexibility in the second year choosing the electives from a very wide offer. It included many practical sessions and visits to the hospital. But sometimes is was difficult to understand the global picture and the purpose of some contents of the program. There were some overlaps. Her favorite course was Health Information and Decision support systems. The collaboration with the other students from different cultures lead sometimes towards cumbersome communication but in the end, it was enriching. Sofia’s golden tip for future students is: Be true to yourself and don’t be afraid of following your goals, even when you get demotivated due to bad scores or difficulties with learning, especially with courses you don’t like but that are mandatory.
Sofia would like to end up applying her knowledge improving people’s lives or investigating in a job that fulfills her and that she is proud of.
Today the results of Fonds Wetenschappelijk Onderzoek – Vlaanderen Mandaten aspirant strategisch basisonderzoek were announced. 196 Mandates were awarded, 13 for VUB and TWO for ETRO. Very well done, Cedric Baijot (prom. Maarten Kuijk) and Salar Tayebi (Prom. Johan Stiens)! Also Delphine Van Laethem, who is doing a PhD with Jeroen Van Schependom will get the scholarship.
Financed by IMEC, CAD for VLSI circuits is launched at ETRO.
4D CT scanners add the dimension of time to three-dimensional images and visualise the movement of the heart in detail. The imec.icon project DIASTOLE, involving VUB, UZ Brussel and imec, is paving the way to safely implement 4D scans in heart surgery.
Researchers from the radiology department of VUB-UZ Brussel developed a model to calculate the radiation dose of 4D scans on the skin, and immediately applied it to draw up a safe protocol. For a usable 4D scan, on the one hand the quality has to be sufficient, on the other hand you want to avoid the radiation dose being too high at certain places on the body. Unlike classic CT scans, a 4D scan repeatedly irradiates the same region of the body, so we need to specifically monitor the dose to the skin.