Group Members

	Álvaro López Paredes Room : PB-H202 E-Mail Alvaro Lopez Paredes is doing his PhD on "Efficient very-wide-area ToF 3D sensing by means of Adaptive Compressive Sensing" within the MSCA-ITN MENELAOS_NT. His aim is to apply adaptive compressive sensing techniques to ToF 3D imaging in order to enhance the accuracy of state-of-the-art ToF sensors as well as to reduce the computational cost derived from the image retrieval, in order to facilitate acquisition, shorten post-processing times and, eventually, increase the applications of this technology to very-wide-areas, long to very long distances and (nearly) real-time operation.
	Faisal Ahmed Room : PB-H009 E-Mail Faisal Ahmed is doing his PhD on "Pseudo-passive indoor ToF 3D sensing exploiting light-based wireless communications infrastructure" within the MSCA-ITN MENELAOS_NT. His aim is to exploit existing wireless communications infrastructure, such as LiFi lamps and VLC modules, as opportunity illuminators for enabling ToF imaging without light emission. This will dramatically decrease the power consumption of ToF cameras and allow them entering new application fields in the context of smart homes, office environments, industrial areas, and vehicles, where VLC infrastructure is being deployed.
	Peyman Fayyaz Shahandashti USC E-Mail Peyman Fayyaz is doing his PhD on "Fabrication of CMOS ToF sensors with 2D/3D capabilities" within the MSCA-ITN MENELAOS_NT. His main research objectives are designing and validating a novel CMOS ToF sensor with 2D/3D capabilities. Additionally, he investigates the feasibility of per-pixel demodulation signal customization. This is expected to yield single-shot ToF cameras with unprecedented measurement diversity, which will be a valuable asset for solving open problems, such as the multi-path interference or 3D sensing in highly-scattering media.
	Adolphe Ndagijimana Room : PB-H105 E-Mail Adolphe Ndagijimana is doing his PhD on the “Compressive sensing and Imaging using TeraHertz” within the MSCA-ITN MENELAOS_NT. The main objective of the project is the improvement of THz imaging systems by applying compressive sensing techniques. This implies both the implementation of novel sensing schemes and the corresponding algorithmic counterpart for image reconstruction. On the hardware front, optical switching will be leveraged for attaining spatial modulation, while metamaterials design is also a prospective research topic.
	Quiang Song Room : PB-H009 E-Mail Qiang Song is doing his Master thesis on "Performance Evaluation of High-Resolution Time-of-Flight Cameras". His aim is to compare the key parameters of state-of-the-art Time-of-Flight (ToF) cameras in order to unveil the strengths and shortcomings of the available technology. This will be a valuable asset in multiple application fields, ranging from human-computer interaction to robotics and autonomous systems.
	Shuo Li Room : PB-H015 E-Mail Shuo Li is doing his study project on "Servo-operated focus and aperture control for a novel ToF camera". The objectives to be used in this camera are fully manual and, therefore, his aim is to develop an electro-mechanical system for computer-controlled adjustment of the manual focus and aperture controls of the lens.
	Zhibin Liu E-Mail Zhibin Liu is doing his student work "Evaluation and compensation of the effect of dirt on ToF 3D imaging systems". His goal is to assess to what extent hybrid multi-frequency operation can improve the accuracy of state-of-the-art ToF cameras w.r.t. single frequency measurements. This will shred light on the applicability of ToF cameras in emerging fields, such as autonomous driving and mobile systems operating under extreme conditions.
	Santosh Kumar Kasam E-Mail Santosh Kumar Kasam is a Research Assistant working on “Performance evaluation of state-of-the-art neural network architectures for ToF-based material classification”. The aim is to predict the material of the object as accurately as possible from NIR images acquired using a PMD ToF camera. He intends to achieve this by a) properly accounting for the effect of signal amplitude, depth, and angle of incidence, b) generating novel features, c) evaluating different state-of-the-art neural network architectures used in computer vision. The resulting accuracy will unveil the feasibility of the camera for commercial applications.
	Varun Kalasapura Lokesh E-Mail Varun Kalasapura Lokesh is doing his student work on “High frequency Illumination for Time-of-flight Cameras”. The illumination frequency is linked to the depth resolution that these 3D imaging systems can achieve and, thus, is a parameter of paramount importance. The objective of his work is to evaluate the behaviour of state-of-the-art NIR VCSELs and compare it to that of existing illumination modules featuring NIR LEDs. Among other critical parameters, the effective rising and falling times, the modulation bandwidth, and the time delay introduced by the emitter and associated driving circuitry and cables will be experimentally evaluated. The results will unveil to which extent faster illumination modules can help to pave the way to multi-frequency operation and to improve the performance of ToF cameras.
	Ravibhai Dhola Room : PB-H009 E-Mail Ravibhai Dhola is doing his student work on "Spatial Superresolution for Time-of-Flight cameras using compressive sensing. " The main goal of his work is to build an optical setup for a multi-single-pixel Time-of-Flight (ToF) camera, using a high-resolution Spatial Light Modulator (SLM) and a Photonic Mixer Device(PMD) ToF pixel array of lower resolution. The latter will be used to superimpose custom patterns onto the imaged scene before reprojection onto the ToF array. Using several patterns on a per-pixel basis, measurements will be obtained that allow for superresolved ToF imaging using compressed sensing. This project is of relevance due to the resolution gap between Tof pixel arrays and conventional cameras.
	Rajababu Udainarayan Singh E-Mail Rajababu Udainarayan Singh is doing his student work on "Time-of-flight based material imaging using three dimensional deep neural networks on spatial neighbourhoods of pixels." His aim is to use multi-frequency ToF data to classify materials on a super-pixel basis using spatial features defined over the neighbourhood of each pixel. This will be accomplished using three-dimensional deep neural networks to process data acquired with a PMD-based ToF camera operated in multi-frequency mode. The results of this research are expected to constitute a step forward towards robust material imaging, unleashing new application domains for ToF cameras.
	Sanhita Guha E-Mail Sanhita Guha is doing her PhD on “Adaptive Compressed Sensing methods for more efficient radar detection and localization”, within the MSCA-ITN MENELAOS_NT. The goal of her work is to explore compressed sensing methods for radar band fusion, leading to better target localization. This will alleviate the effects of spectrum congestion and allow the use of existing narrow-band radars to achieve better range resolution.
	Saravanan Nagesh E-Mail Saravanan Nagesh is working on his PhD titled “Coded waveform design for Compressed Sensing MIMO radar systems” within the MSCA-ITN MENELAOS_NT. His aim is to design “sensing matrices” with low mutual coherence, which are preferred when using Compressed Sensing (CS) algorithms for target parameter estimation or scene reconstruction using MIMO radar systems. The design of the sensing matrix directly effects the MIMO system parameters, such waveform sequences which in turn aids in improved efficiency and reliability of the considered MIMO system. Outputs of his research work find applications in the area of, e.g., autonomous driving. He currently works at the Cognitive Radar department at the “Fraunhofer-Institut für Hochfrequenzphysik und Radartechnik”, FHR.
	Mobina Kaymasi E-Mail Mobina Keymasi is pursuing her PhD on “Goal-directed compressive radar imaging” within the MSCA-ITN MENELAOS_NT. The main objective of the project is to develop new compressive radar imaging methods that explicitly consider and adapt to the final decision-making goals to preserve and enhance features in the scene for high-level goals such as object recognition or scene classification. This implies the implementation of both Artificial Intelligence and Compressive Sensing techniques on Earth Observation data.
	Navtegh Gill Room : PB-H105 E-Mail Navtegh Singh Gill is a visiting researcher doing research on “Creating and accelerating multi-dimensional deep learning models for the classification of different materials on a per-pixel basis using Time-of-Flight (ToF) data”. This will be accomplished by using various methods such as Principal Component Analysis, Quantisation, Network Pruning, etc. The results of this research are expected to reduce the complexity of current neural network models and significantly decrease the training and inferencing time required for the classification of materials using ToF data.
	Bharat Lal Room : PB-H017 E-Mail Bharat Lal is a visiting PhD researcher at ZESS, where his research focuses on the application of Compressive Sensing (CS) and Compressive Learning (CL) techniques in healthcare systems. His primary objective is to address the challenges associated with managing large volumes of physiological data and reducing power consumption in wireless healthcare devices for continuous patient monitoring. Bharat's work specifically emphasizes the utilization of CS and CL for the analysis of physiological signals and the extraction of diagnosis-relevant features, with a special focus on electrocardiography (ECG), electromyography (EMG), and electroencephalography (EEG). Furthermore, he explores the implementation of these techniques in hardware platforms such as FPGA to effectively overcome the aforementioned challenges.
	Muhammad Amjad Iqbal Room : PB-H105 E-Mail Muhammad Amjad Iqbal is pursuing his Ph.D. under the MSCA-ITN MENELAOS_NT, with a thesis titled “Sparse Reconstruction for High-Resolution Inverse SAR Imaging.” The objective of his research is target detection using complex radar data. His work involves signal and image processing of SAR and ISAR data for information retrieval. His aim is to utilize low-resolution data to extract Doppler features in high-resolution images. His main contribution is coastline detection using Doppler images. During his research stay at ZESS, he focused on sparsity-driven ISAR imaging.
	Omid Ghozatlou Room : PB-H105 E-Mail Omid Ghozatlou is currently working toward the Ph.D. degree in the field of “Learning with adversarial samples for Earth Observation (EO) images” within the MSCA-ITN MENELAOS_NT. His project aims to provide solutions for deep learning for EO multi-spectral images in the presence of naturally occurring adversarial samples and also considering their physical nature and models. His current research interests include physics-aware artificial intelligence, SAR image generation, and deep active learning.

Aktualisiert via XIMS am 27.7.2023 von G. Basten