TY - CHAP AU - Dos Santos Melício, Bruno Carlos AU - Xiang, Linyun AU - Dillon, Emily AU - Soorya, Latha AU - Chetouani, Mohamed AU - Sarkany, Andras AU - Kun, Peter AU - Fenech, Kristian AU - Lőrincz, András ED - André, Elisabeth ED - Chetouani, Mohamed TI - Composite AI for Behavior Analysis in Social Interactions T2 - 25th International Conference on Multimodal Interaction PB - Association for Computing Machinery (ACM) CY - New York, New York SN - 9798400703218 PY - 2023 SP - 389 EP - 397 PG - 9 DO - 10.1145/3610661.3616237 UR - https://m2.mtmt.hu/api/publication/34201511 ID - 34201511 LA - English DB - MTMT ER - TY - JOUR AU - Fodor, Ádám AU - Fenech, Kristian AU - Lőrincz, András TI - BlinkLinMulT: Transformer-Based Eye Blink Detection JF - JOURNAL OF IMAGING J2 - J IMAGING VL - 9 PY - 2023 IS - 10 SN - 2313-433X UR - https://m2.mtmt.hu/api/publication/34159345 ID - 34159345 LA - English DB - MTMT ER - TY - JOUR AU - Kopácsi, László AU - Baffy, Benjámin AU - Baranyi, Gábor AU - Skaf, Joul AU - Sörös, Gábor AU - Szeier, Szilvia AU - Lőrincz, András AU - Sonntag, Daniel TI - Cross-Viewpoint Semantic Mapping: Integrating Human and Robot Perspectives for Improved 3D Semantic Reconstruction JF - SENSORS J2 - SENSORS-BASEL VL - 23 PY - 2023 IS - 11 SN - 1424-8220 DO - 10.3390/s23115126 UR - https://m2.mtmt.hu/api/publication/34045337 ID - 34045337 AB - Allocentric semantic 3D maps are highly useful for a variety of human–machine interaction related tasks since egocentric viewpoints can be derived by the machine for the human partner. Class labels and map interpretations, however, may differ or could be missing for the participants due to the different perspectives. Particularly, when considering the viewpoint of a small robot, which significantly differs from the viewpoint of a human. In order to overcome this issue, and to establish common ground, we extend an existing real-time 3D semantic reconstruction pipeline with semantic matching across human and robot viewpoints. We use deep recognition networks, which usually perform well from higher (i.e., human) viewpoints but are inferior from lower viewpoints, such as that of a small robot. We propose several approaches for acquiring semantic labels for images taken from unusual perspectives. We start with a partial 3D semantic reconstruction from the human perspective that we transfer and adapt to the small robot’s perspective using superpixel segmentation and the geometry of the surroundings. The quality of the reconstruction is evaluated in the Habitat simulator and a real environment using a robot car with an RGBD camera. We show that the proposed approach provides high-quality semantic segmentation from the robot’s perspective, with accuracy comparable to the original one. In addition, we exploit the gained information and improve the recognition performance of the deep network for the lower viewpoints and show that the small robot alone is capable of generating high-quality semantic maps for the human partner. The computations are close to real-time, so the approach enables interactive applications. LA - English DB - MTMT ER - TY - JOUR AU - Szijártó, Ádám AU - Somfai, Ellák AU - Lőrincz, András TI - Design of a Machine Learning System to Predict the Thickness of a Melanoma Lesion in a Non-Invasive Way from Dermoscopic Images JF - Healthcare Informatics Research J2 - Healthcare Informatics Research VL - 29 PY - 2023 IS - 2 SP - 112 EP - 119 PG - 8 SN - 2093-369X DO - 10.4258/hir.2023.29.2.112 UR - https://m2.mtmt.hu/api/publication/33947821 ID - 33947821 N1 - Export Date: 08 March 2024 LA - English DB - MTMT ER - TY - JOUR AU - Somfai, Ellák AU - Baffy, B. AU - Fenech, Kristian AU - Hosszú, R. AU - Korózs, D. AU - Pólik, Marcell AU - Sárdy, Miklós AU - Lőrincz, András TI - Handling dataset dependence with model ensembles for skin lesion classification from dermoscopic and clinical images JF - INTERNATIONAL JOURNAL OF IMAGING SYSTEMS AND TECHNOLOGY J2 - INT J IMAG SYST TECH VL - 33 PY - 2023 IS - 2 SP - 556 EP - 571 PG - 16 SN - 0899-9457 DO - 10.1002/ima.22827 UR - https://m2.mtmt.hu/api/publication/33416651 ID - 33416651 N1 - Export Date: 08 March 2024; CODEN: IJITE LA - English DB - MTMT ER - TY - CHAP AU - Bóna, Judit AU - Gosztolya, Gábor AU - Hoffmann, Ildikó AU - Klivényi, Péter AU - Tóth, Alinka AU - Svindt, Veronika AU - Tóth, László AU - Lőrincz, András ED - Dobrić, Arnalda ED - Liker, Marko TI - Temporal variables of speech in Parkinson’s Disease in three spontaneous speaking tasks T2 - Book of Abstracts : The 11th scientific conference with international participation Speech Research, Faculty of Humanities and Social Sciences, Zagreb, Croatia, December 8 - 10 2022 PB - Hrvatsko filološko društvo CY - Zágráb SN - 9789532961935 PY - 2022 SP - 28 EP - 29 PG - 2 UR - https://m2.mtmt.hu/api/publication/33576741 ID - 33576741 LA - English DB - MTMT ER - TY - CHAP AU - Szeghy, David AU - Aslan, Mahmoud AU - Fóthi, Áron AU - Meszaros, Balazs AU - Milacski, Zoltán Ádám AU - Lőrincz, András ED - Fred, A ED - Sansone, C ED - Gusikhin, O ED - Madani, K TI - Structural Extensions of Basis Pursuit: Guarantees on Adversarial Robustness T2 - DELTA: PROCEEDINGS OF THE 3RD INTERNATIONAL CONFERENCE ON DEEP LEARNING THEORY AND APPLICATIONS PB - SciTePress CY - Setubal SN - 9789897585845 PY - 2022 SP - 77 EP - 85 PG - 9 DO - 10.5220/0011138900003277 UR - https://m2.mtmt.hu/api/publication/33305314 ID - 33305314 AB - While deep neural networks are sensitive to adversarial noise, sparse coding using the Basis Pursuit (BP) method is robust against such attacks, including its multi-layer extensions. We prove that the stability theorem of BP holds upon the following generalizations: (i) the regularization procedure can be separated into disjoint groups with different weights, (ii) neurons or full layers may form groups, and (iii) the regularizer takes various generalized forms of the l(1) norm. This result provides the proof for the architectural generalizations of (Cazenavette et al., 2021) including (iv) an approximation of the complete architecture as a shallow sparse coding network. Due to this approximation, we settled to experimenting with shallow networks and studied their robustness against the Iterative Fast Gradient Sign Method on a synthetic dataset and MNIST. We introduce classification based on the l(2) norms of the groups and show numerically that it can be accurate and offers considerable speedups. In this family, linear transformer shows the best performance. Based on the theoretical results and the numerical simulations, we highlight numerical matters that may improve performance further. The proofs of our theorems can be found in the supplementary material*. LA - English DB - MTMT ER - TY - GEN AU - Fenech, Kristian AU - Fodor, Ádám AU - Bergeron, Sean Paul AU - Saboundji, Rachid Rhyad AU - Oertel, Catharine AU - Lőrincz, András TI - Perceived personality state estimation in dyadic and small group interaction with deep learning methods VL - abs/2211.04979 PY - 2022 UR - https://m2.mtmt.hu/api/publication/33272087 ID - 33272087 LA - English DB - MTMT ER - TY - CHAP AU - Tsfasman, Maria AU - Fenech, Kristian AU - Tarvirdians, Morita AU - Lőrincz, András AU - Jonker, Catholijn AU - Oertel, Catharine TI - Towards creating a conversational memory for long-term meeting support: predicting memorable moments in multi-party conversations through eye-gaze T2 - INTERNATIONAL CONFERENCE ON MULTIMODAL INTERACTION PB - Association for Computing Machinery (ACM) CY - New York, New York SN - 9781450393904 PY - 2022 SP - 94 EP - 104 PG - 11 DO - 10.1145/3536221.3556613 UR - https://m2.mtmt.hu/api/publication/33214815 ID - 33214815 LA - English DB - MTMT ER - TY - GEN AU - Éva, Mariczáné Csanádi* AU - Fóthi, Ábel AU - Péter, Pollner AU - Lőrincz, András AU - Bálint, Kovács TI - Opportunities and Limitations in the Analysis of Autistic Phenotypes - A Conceptual Review PY - 2022 UR - https://m2.mtmt.hu/api/publication/33113731 ID - 33113731 LA - English DB - MTMT ER -