Research
Research fields
2D Sensors based Perception
• 2D Object Detection, Recognition and Tracking;	
• 2D Semantic, Instance and Panoptic Segmentation;
• 2D Video Panoptic Segmentation;
• Camera calibration;
• Binocular and trinocular stereovision;
• Structure from Motion, Monocular Metric Depth Estimation;
• Design and implementation of hardware solutions for stereovision using FPGA technology.
3D Sensors based Perception
• LiDAR and RADAR Data Preprocessing;
• 3D Point Cloud based Perception;
• 3D Object Detection, Recognition and Tracking;	
• 3D Semantic, Instance and Panoptic Segmentation;
• Geometry and Semantic Fusion and Enhanced Perception Solutions;
• Environment Representation; 
• Risk Assessment, Planning; 
• Collision Avoidance, Navigation.
Stereovision Based Perception for Driving Assistance Systems and Autonomous Driving 
• 3D lane detection;
• 3D objects detection and tracking;
• 3D objects classification;
• Pedestrian detection;
• 3D structured/unstructured environment modeling;
• Risk assessment.
Applications
• Perception based advanced driving assistance applications;
• 3D Perception system for autonomous fork lifts;
• Holistic perception and scene understanding for autonomous driving;
• Real time percetion for intelligent drones.
Intelligent Space Surveilance
• Optical surveillance of orbiting objects.
Medical Image Processing
• Texture based detection and classification of diffuse and focal illness from Ultrasound Images;
• DICOM infrastructure implementation;
• Structured reporting for medical images;
• Advanced methods for segmentation and classification of anatomical structures from multimodal medical images.
Research projects
Recent research projects
"HRIA – Romanian Artificial Intelligence Hub", Program POCIDIF, Ministry of Investments and European Projects (2025–2029)
"SenseMaking – Distributed Sensemaking", funded by Lockheed Martin (2022–2024)
"SEPCA - Integrated Semantic Visual Perception and Control for Autonomous Systems", grant funded by Romanian Ministry of Education and Research, code PN-III-P4-ID-PCCF-2016-0180, (2018-2022)
"DeepSense - Advanced environment perception techniques based on Deep Learning and probabilistic estimators", grant funded by Romanian Ministry of Education and Research, code PN-III-P1-1.1-TE-2016-0440 (2018-2020)
"MULTISPECT - Multispectral environment perception by fusion of 2D and 3D sensorial data from the visible and infrared spectrum", grant funded by Romanian Ministry of Education and Research, code PN-III-P4-ID-PCE-2016-0727(2017-2019)
Older Projects
- Study and implementation of a hardware device for real-time convolution. Beneficiary: Tolmi Electronics Ltd. Ireland
 - Study of the computational matching strategies based on fitting the model to symbolic structures. Implementation of an experimental model-based recognition system for industrial objects, with application in robotics. Beneficiary: Romanian Ministry of Education and Research
 - Study and implementation of a method for symbolic representation of knowledge in model-based recognition. Beneficiary: Romanian Ministry of Education and Research
 - Model-based industrial objects recognition with application in robotics. Beneficiary: Romanian Ministry of Education and Research
 - Pattern recognition methods based on descriptors and syntactic descriptions. Beneficiary: ITC Cluj Napoca
 


