Richard Capraru

: Training deep learning perception networks to understand the baseline signature of rain or fog, making anomalous laser injections stand out.

: Capraru's research forces developers to treat weather not just as an operational hurdle, but as a severe cybersecurity attack vector.

During his undergraduate years, Capraru was honored as a , a prestigious recognition that champions research and leadership development. This opportunity allowed him to work with the UCL Radar Research Group , giving him early, hands-on experience in one of the foundational technologies for sensing the environment. This initial exposure has powerfully shaped his career, as radar later became a recurring theme in his research.

"Exploring deep transfer learning interference classification on neural style transfer generated synthetic SAR datasets" (2022). richard capraru

He explores the performance of LiDAR vision systems in self-driving cars during heavy rain. His work highlights how rain can be leveraged by attackers to create "ghost objects" or hide real obstacles with a reduced attack budget.

After completing his degree, Capraru continued his research journey in Singapore. He is currently a at the School of Electrical and Electronic Engineering (EEE) at Nanyang Technological University (NTU) . In parallel, he is also a student at Singapore's A*STAR Institute for Infocomm Research (I²R) . He holds a SINGA scholarship from A*STAR, a testament to his status as a rising scientific talent. This dual role allows him to bridge advanced academic theory with applied research in a national laboratory setting.

In his residential work, Capraru often plays with the concept of the "edited view." He understands that the most expensive view is not the one that is wide open, but the one that is framed. By using architectural elements to crop the landscape—turning a window into a living painting—he forces the occupant to engage with the environment in a specific, intentional way. It is a directorial move, positioning the resident not just as a dweller, but as an audience member to their own life. : Training deep learning perception networks to understand

Richard Capraru, Research Assistant and Student, UCL. I am a/an: Alum: Undergraduate Leadership & Research Programme. Laidlaw Scholars Network

"Upsampling Data Challenge: Object-Aware Approach for 3D Object Detection in Rain" (2023).

Richard Capraru is a researcher and academic specializing in the intersection of , cybersecurity , and signal processing . He is currently a PhD candidate at Nanyang Technological University (NTU) in Singapore and the Institute for Infocomm Research (I2R) at A*STAR. Academic Background and Education This opportunity allowed him to work with the

: He has investigated the security of autonomous driving systems, specifically focusing on LiDAR spoofing and real-time attacks, such as "GhostLite," which explores data minimization for high-speed sensor interference. Gesture Recognition

This public link is valid for 7 days and shares a thread, including any personal information you added. This link or copies made by others cannot be deleted. If you share with third parties, their policies apply. Can’t copy the link right now. Try again later.