“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.
“Imaging & Machine Vision Europe“‘s Abigail Williams speaks to scientists tracking marine plastic using satellite spectral imagery. For ETRO, Prof Jonathan C-W Chan was interviewed. Below you can find his remarks.
Professor Jonathan Chan, noted that the ideal sensor to detect marine plastic should possess more spectral measurement capabilities at SWIR wavelengths.
Chan is working on the Muss2 project, which is using spectral and spatial enhancement methods to generate simulated Earth orbit hyperspectral shortwave infrared images and data from the Copernicus Sentinel 2 satellite using spectral response function modelling.
Hyperspectral images taken from Earth orbit are not always available and their coverage is not as large as conventional missions, such as Landsat and Sentinel. To overcome these limitations, Chan said the Muss2 team will apply a sparse theory-based method to enhance multispectral images from the Sentinel 2 satellite. The expected results are what he described as synthetic Sentinel 2 hyperspectral images at a spatial resolution of 10m, with the same coverage as Sentinel 2 multispectral images.
‘So far, we have been able to generate such images based on Hyperion and Prisma spectral configurations, [and] quantitative assessments are promising,’ he said. ‘In addition, we apply a deep learning-based method for the spatial enhancement of Prisma images to assist in the detection of smaller objects.’
Chan added: ‘All the data inputs for the project are from open sources, including Sentinel 2, Hyperion and Prisma. The launch of Prisma in 2019 was a major milestone, and soon EnMap will be launched. These are full stretch hyperspectral images at 400-2,500nm with hundreds of spectral bands and represent the most powerful Earth orbit data source for environmental monitoring. The potential is not yet fully understood nor exploited, and scientific communities are just beginning to understand all the pre-processing protocols and data quality issues.’
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.
Fawaz Samani received the FWO Aspirant strategisch basisonderzoek mandate with the project “Towards Human Friendly Explanations”
The framework agreement with Omnia facilitates VUB’s travel policy. Omnia’s booking platform, where employees can book all their business trips, was aligned with the Travel ABC. For example, a new carbon offset tool was added, and the platform displays whether your destination is within train distance. The platform is accessible with your VUB login, but you can also book through the Omnia office in Brussels at the lowest price.
Via email: Brussels@omniatravel.be (for general support)
By phone: 02 645 56 10 (Mon-Fri, 9.00-18.00)
The guide for booking business trips can be found in ServiceNow or can be viewed in video format.
We can also book your trips via Voyages Plus: Jesus Voyages Plus firstname.lastname@example.org
PHD Day 22/2/24
Selene De Sutter started her PhD at ETRO in 2019. Since childhood she suffers from Diabetes type 1 and she received an insulin pump for treating this. The choice of following the Master Biomedical Engineering was obvious, so that she could develop technologies that would help people with medical conditions, like the insulin pump did for her. If Selene did not choose for am Engineering training you could have met her on stage, as she was hesitating to follow a Jazz training. Now she combines Engineering Sciences with music as a hobby.
The program creates lots of possibilities so that you have a lot of options. It starts general and you can choose options along the go according to your interests. It does take some perseverance though!
Biomedical Imaging was the course Selene liked most in the master Biomedical Engineering. A very visual kind of programming that showed its usefulness in practice immediately. It also became her topic for her PhD which is her next big goal. What will happen after the PhD is not clear yet, but she wants to work on something useful and enjoy what she is doing.
A golden tip from Selene: Working hard is important, but don’t forget to enjoy yourself at the university!