“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)
Bachelor students in their last year of study can apply for the program with the condition to successfully finish their Bachelor studies before the start of the academic year.
On March 29th 2024 at 17::00, Jurgen Vandendriessche will defend their PhD entitled “TOWARDS SMART ACOUSTIC CAMERAS FOR SIMULTANEOUS SOUND LOCALIZATION AND RECOGNITION”.
Everybody is invited to attend the presentation in room I.0.01, or digitally via this link.
Acoustic cameras are devices that visualize sound by utilizing an array of microphones. The signal from each microphone is combined using a beamforming algorithm to generate an acoustic heatmap or acoustic image. These beamforming algorithms tend to have a high computational cost, which increases with the number of microphones. The combination of a high number of Input/Output (I/O) requirements for the microphones combined with the high amount of parallel computations makes Field Programmable Gate Arrays (FPGAs) very suitable for processing the signals from these microphone arrays. FPGAs have a low power consumption, which makes them especially viable when targeting battery powered devices such as handheld acoustic cameras or nodes in a sensor network. Despite the high computational power per watt of FPGAs, satisfying real time scenarios still present a challenge, especially when targeting acoustic images with a higher resolution. To overcome this challenge, a multi-mode acoustic camera has been developed. The camera supports multiple modes depending on the task at hand. To satisfy the real time requirement for each mode, the resolution of the acoustic heatmap can be adapted. A second limitation of the existing acoustic cameras is the identification of the type of sound, which commonly requires human expertise to recognize and profile the sound.
In recent years, deep learning, a form of Artificial intelligence (AI), has shown promising results towards the task of sound recognition by using Convolutional Neural Networks (CNNs). However, most of the research focuses on improving the accuracy of such models without considering the limitations one encounters when deploying such a model on resource constrained devices such as FPGAs. FPGAs are used nowadays for embedding deep learning inference, mainly using two architectures. One type of architecture uses a general-purpose soft-core inside the Programmable Logic (PL) of the FPGA. On the other hand, there are also dataflow-based architectures that translate each layer in a CNN to a functional block in the PL. Embedding these CNNs for inference on FPGAs is not a trivial task and come with trade-offs in terms of resource consumption, accuracy, supported layers,… These two architectures are compared against other embedded solutions such as Google’s edge Tensor Processing Unit (TPU) and a Raspberry Pi (RPi) to find the best fit for acoustic cameras. Acoustic cameras are targeted in this instance because they can identify the location of a sound source, which is not possible when using one microphone. Furthermore, existing beamforming techniques such as delay-and-sum reconstruct audio signals, which can be used for audio classification tasks.
Two extra assistants join ETRO: Joris Lievens and Jan Cornelis (electronic circuits and instrumentation) – Prototype PCB manufacturing facility. Bulky instruments based on 74xx MSI ICs are designed: e.g. calculator, noise generator, datalogger with multipen printer, …
Gisteren hebben we tijdens een zonovergoten avond het ecosysteem van Green Energy Park vzw en FACT – VUB samengebracht voor een interessant netwerkevent.
Een inspirerende uitleg over onze ambities, de visie van Siemens over het belang van intense samenwerking, overtuigende pitches van de onderzoeksgroepen van FACT – VUB, een infosessie over het Interreg-project EMR H2-HUBs, verschillende demo’s … een mooi programma waar iedereen zijn gading in kon vinden.
De avond werd afgesloten onder een stralende zon waardoor het netwerken net dat tikkeltje aangenamer was.
Bedankt aan onze sprekers Inne Peersman, Siperius Yves, Maarten Messagie, Cedric De Cauwer, Gabriela Barrera, Md Sazzad Hosen, Christof Devriendt, Aurélie Bellemans, Lukas Berger, Tom Van Assche, Johan Stiens, Hubert Rahier en onze demostandhouders Priva, Fluvius, Allspark, Flux50, ABB, EVERGi, ETRO VUB, Physical Chemistry and Polymer Science en VUB Architectural Engineering.
From lab to impact – 20th May – 27th May – 3rd June
Are you a researcher looking to amplify the real-world impact of your work? Join us for From lab to impact, a series of 3 info sessions designed to help you imagine what that future could look like.
This series is hosted by StartLab.Brussels and VUB and brings you inspiring real-life stories, plenty of practical insights and the opportunity to connect with researchers who have successfully made the leap.
Session 1: Real-world applications are born in the lab. Real stories of research that SPIN-off.
Florian Myter, former VUB PhD researcher and Co-founder of weave.ly
Jens De Vos, former VUB PhD researcher and Head of Alliance and Program Management of Precirix
🗓️ May 20th, 2025 – 18:00 – 20:30 (Including networking drinks)
đź“Ť Learning & Innovation (LIC) Center, VUB
đź”— More info & registration
Session 2 (given by your own ETRO colleagues): Funded to make impact. Real stories of innovation mandates in action.
Pieter Thomas Boonen, Postdoctoral researcher – Project PADFLOW
Benyameen KEELSON, Postdoctoral researcher – project Karma
Taylor Frantz, Postdoctoral researcher – Project Sara
🗓️ May 27th, 2025 – 12:00 – 14:00 (sandwich lunch included)
đź“Ť Learning & Innovation (LIC) Center, VUB
đź”— More info & registration
Session 3: Building a team to create impact. Real stories of turning solo ideas into team successes.
Speakers will be confirmed soon.
🗓️ June 3rd, 2025 – 12:00 – 14:00 (sandwich lunch included)
đź“Ť Learning & Innovation (LIC) Center, VUB
đź”— More info & registration
HealthTech TouchPoints event (17/10/2024): VUB and UZ Brussel showcased their HealthTech expertise to companies.
ETRO did a pitch and had a demo booth at the matchmaking fair after the event.
More background info LinkedIn posts:
- https://www.linkedin.com/posts/umcor_vubhealthcampus-umcor-umcorcommunity-activity-7254163276907909120-0AyH?utm_source=share&utm_medium=member_desktop
- https://www.linkedin.com/posts/medvia-be_healthtech-touchpoints-healthinnovation-activity-7254469077698539520-B90z?utm_source=share&utm_medium=member_desktop
