Univerza na Primorskem Fakulteta za matematiko, naravoslovje in informacijske tehnologije
-->
SI | EN

ponedeljek, 23. avgust 2021 Nuwan T. ATTYGALLE: Missing Interface - Gesture Recognition on Physical Objects using Radar Sensing

V ponedeljek, 23. avgusta 2021, bo ob 16.00 uri prek spletnih orodij na daljavo izvedeno
predavanje v okviru PONEDELJKOVEGA SEMINARJA RAČUNALNIŠTVA IN INFORMATIKE
Oddelkov za Informacijske znanosti in tehnologije UP FAMNIT in UP IAM.

ČAS/PROSTOR: 23. avgust 2021 ob 16.00 na daljavo

-------------------------------------------------
PREDAVATELJ: Nuwan T. ATTYGALLE
-------------------------------------------------

Nuwan Tharanga Attygalle is a PhD student in Computer Science in the field of Human-Computer interactions and a teaching assistant at the University of Primorska, Faculty of Mathematics, Natural Sciences and Information Technologies. He is also a member of the HICUP Lab at the Department of Information Sciences and Technologies.

------------------------------------------------------------------------------------------------------------------------
NASLOV: Missing Interface: Gesture Recognition on Physical Objects using Radar Sensing
------------------------------------------------------------------------------------------------------------------------

POVZETEK:

Physical objects are usually not designed with interaction capabilities to control digital content. Nevertheless, they provide an untapped source for interactions since every object could be used to control our digital lives. We call this the missing interface problem: Instead of embedding computational capacity into objects, we can simply detect users’ gestures on them. However, gesture detection on such unmodified objects so far has been limited in spatial resolution and detection fidelity. To address this gap, we conducted research on micro-gesture detection on physical objects based on Google Soli’s radar sensor. We introduce two novel deep learning architectures to process range-Doppler images, namely a 3-dimensional convolutional neural network (Conv3D) and a spectrogram-based ConvNet. The results show that our architectures enable robust on-object gesture detection, achieving accuracy around 94% for a 5-gesture set, surpassing previous state-of-the-art performance results by up to 39%. We also show that the decibel (dB) Doppler range setting has a significant effect on system performance, as accuracy can vary up to 20% across the dB range. As a result, we provide guidelines on how to best calibrate the radar sensor.

---------------------------------------------------------------------------------------------------

Predavanje bo potekalo v angleškem jeziku prek spletnega orodja Zoom.
Do predavanja dostopate tako, da se povežete prek sledeče povezave:

https://upr-si.zoom.us/j/297328207

Vabljeni!