TY - JOUR AU - Kovács, Lóránt AU - Bódis, Balázs Márk AU - Benedek, Csaba TI - LidPose: Real-Time 3D Human Pose Estimation in Sparse Lidar Point Clouds with Non-Repetitive Circular Scanning Pattern JF - SENSORS J2 - SENSORS-BASEL VL - 24 PY - 2024 IS - 11 PG - 25 SN - 1424-8220 DO - 10.3390/s24113427 UR - https://m2.mtmt.hu/api/publication/34881244 ID - 34881244 AB - In this paper, we propose a novel, vision-transformer-based end-to-end pose estimation method, LidPose, for real-time human skeleton estimation in non-repetitive circular scanning (NRCS) lidar point clouds. Building on the ViTPose architecture, we introduce novel adaptations to address the unique properties of NRCS lidars, namely, the sparsity and unusual rosetta-like scanning pattern. The proposed method addresses a common issue of NRCS lidar-based perception, namely, the sparsity of the measurement, which needs balancing between the spatial and temporal resolution of the recorded data for efficient analysis of various phenomena. LidPose utilizes foreground and background segmentation techniques for the NRCS lidar sensor to select a region of interest (RoI), making LidPose a complete end-to-end approach to moving pedestrian detection and skeleton fitting from raw NRCS lidar measurement sequences captured by a static sensor for surveillance scenarios. To evaluate the method, we have created a novel, real-world, multi-modal dataset, containing camera images and lidar point clouds from a Livox Avia sensor, with annotated 2D and 3D human skeleton ground truth. LA - English DB - MTMT ER - TY - CONF AU - Kovács, Lóránt AU - Kégl, Marcell AU - Benedek, Csaba TI - Real-time foreground segmentation for surveillance applications in sequences from a Non-repetitive Circular Scanning Lidar T2 - Proceedings of KEPAF 2023: Képfeldolgozók és Alakfelismerők társaságának 14. konferenciája PY - 2023 UR - https://m2.mtmt.hu/api/publication/34693946 ID - 34693946 LA - English DB - MTMT ER - TY - CONF AU - Kégl, Marcell AU - Kovács, Lóránt AU - Benedek, Csaba TI - Real-Time Surveillance Application for NRCS Lidar Sequences T2 - Proceedings of KEPAF 2023: Képfeldolgozók és Alakfelismerők társaságának 14. konferenciája PY - 2023 UR - https://m2.mtmt.hu/api/publication/34693938 ID - 34693938 LA - English DB - MTMT ER - TY - JOUR AU - Kovács, Lóránt AU - Kégl, Marcell AU - Benedek, Csaba TI - Real-time foreground segmentation for surveillance applications in NRCS lidar sequences JF - INTERNATIONAL ARCHIVES OF PHOTOGRAMMETRY AND REMOTE SENSING (2002-) J2 - ISPRS (2002-) VL - 43 PY - 2022 IS - B1-2022 SP - 45 EP - 51 PG - 7 SN - 1682-1750 DO - 10.5194/isprs-archives-XLIII-B1-2022-45-2022 UR - https://m2.mtmt.hu/api/publication/32925042 ID - 32925042 LA - English DB - MTMT ER - TY - JOUR AU - Kovács, Lóránt TI - Change detection in unregistered 3D point clouds JF - PHD PROCEEDINGS ANNUAL ISSUES OF THE DOCTORAL SCHOOL FACULTY OF INFORMATION TECHNOLOGY AND BIONICS J2 - PHD PROC PPKE IT VL - 16 PY - 2021 SP - 67 SN - 2064-7271 UR - https://m2.mtmt.hu/api/publication/32660821 ID - 32660821 LA - English DB - MTMT ER - TY - JOUR AU - Nagy, Balázs AU - Kovács, Lóránt AU - Benedek, Csaba TI - ChangeGAN: A deep network for change detection in coarsely registered point clouds JF - IEEE ROBOTICS AND AUTOMATION LETTERS J2 - IEEE ROBOT AUTOM LETT VL - 6 PY - 2021 IS - 4 SP - 8277 EP - 8284 PG - 8 SN - 2377-3766 DO - 10.1109/LRA.2021.3105721 UR - https://m2.mtmt.hu/api/publication/32150436 ID - 32150436 N1 - Funding Agency and Grant Number: National Research Development and Innovation (NRDI) Office within Autonomous Systems National Laboratory; Artificial Intelligence National Laboratory programs; NRDI Fund (OTKA) [K-120233, 2018-2.1.3-EUREKA-2018-00032]; Szechenyi 2020 Program [EFOP-3.6.3-VEKOP-16-2017-00002]; New National Excellence Program of the Ministry for Innovation, and Technology [UNKP20-3, UNKP-20-4] Funding text: This work was supported in part by the National Research Development and Innovation (NRDI) Office within the frameworks of the Autonomous Systems National Laboratory, in part by the Artificial Intelligence National Laboratory programs, in part by the NRDI Fund (OTKA) under Grants K-120233, and 2018-2.1.3-EUREKA-2018-00032, in part by the Szechenyi 2020 Program under Grant EFOP-3.6.3-VEKOP-16-2017-00002, and in part by the UNKP20-3, and UNKP-20-4 New National Excellence Program of the Ministry for Innovation, and Technology. LA - English DB - MTMT ER - TY - JOUR AU - Kovács, Lóránt TI - Change detection in Lidar point clouds JF - PHD PROCEEDINGS ANNUAL ISSUES OF THE DOCTORAL SCHOOL FACULTY OF INFORMATION TECHNOLOGY AND BIONICS J2 - PHD PROC PPKE IT VL - 15 PY - 2020 SP - 68 SN - 2064-7271 UR - https://m2.mtmt.hu/api/publication/32660860 ID - 32660860 LA - English DB - MTMT ER - TY - CHAP AU - H. Zováthi, Örkény Ádám AU - Kovács, Lóránt AU - Nagy, Balázs AU - Benedek, Csaba TI - Multi-Object Detection in Urban Scenes Utilizing 3D Background Maps and Tracking T2 - 2019 International Conference on Control, Artificial Intelligence, Robotics & Optimization (ICCAIRO) PB - Institute of Electrical and Electronics Engineers (IEEE) CY - Piscataway (NJ) SN - 9781728135724 PY - 2020 SP - 231 EP - 236 PG - 6 DO - 10.1109/ICCAIRO47923.2019.00044 UR - https://m2.mtmt.hu/api/publication/31275586 ID - 31275586 LA - English DB - MTMT ER - TY - JOUR AU - Kovács, Lóránt TI - Challenges in track to track sensor fusion using neural networks JF - PHD PROCEEDINGS ANNUAL ISSUES OF THE DOCTORAL SCHOOL FACULTY OF INFORMATION TECHNOLOGY AND BIONICS J2 - PHD PROC PPKE IT VL - 14 PY - 2019 IS - 2019 SP - 60 SN - 2064-7271 UR - https://m2.mtmt.hu/api/publication/32660893 ID - 32660893 LA - English DB - MTMT ER - TY - JOUR AU - Kovács, Lóránt TI - Challenges in sensor fusion JF - PHD PROCEEDINGS ANNUAL ISSUES OF THE DOCTORAL SCHOOL FACULTY OF INFORMATION TECHNOLOGY AND BIONICS J2 - PHD PROC PPKE IT VL - 13 PY - 2018 SP - 55 SN - 2064-7271 UR - https://m2.mtmt.hu/api/publication/32660940 ID - 32660940 LA - English DB - MTMT ER -