@inproceedings{MTMT:34201511, title = {Composite AI for Behavior Analysis in Social Interactions}, url = {https://m2.mtmt.hu/api/publication/34201511}, author = {Dos Santos Melício, Bruno Carlos and Xiang, Linyun and Dillon, Emily and Soorya, Latha and Chetouani, Mohamed and Sarkany, Andras and Kun, Peter and Fenech, Kristian and Lőrincz, András}, booktitle = {25th International Conference on Multimodal Interaction}, doi = {10.1145/3610661.3616237}, unique-id = {34201511}, year = {2023}, pages = {389-397}, orcid-numbers = {Dos Santos Melício, Bruno Carlos/0000-0002-2839-8992; Xiang, Linyun/0009-0002-5211-3712; Dillon, Emily/0000-0002-9362-7085; Soorya, Latha/0000-0002-0414-4533; Chetouani, Mohamed/0000-0002-2920-4539; Sarkany, Andras/0000-0002-7506-7563; Kun, Peter/0009-0006-3935-887X; Fenech, Kristian/0000-0002-8288-9303; Lőrincz, András/0000-0002-1280-3447} } @article{MTMT:34159345, title = {BlinkLinMulT: Transformer-Based Eye Blink Detection}, url = {https://m2.mtmt.hu/api/publication/34159345}, author = {Fodor, Ádám and Fenech, Kristian and Lőrincz, András}, journal-iso = {J IMAGING}, journal = {JOURNAL OF IMAGING}, volume = {9}, unique-id = {34159345}, issn = {2313-433X}, year = {2023}, orcid-numbers = {Fenech, Kristian/0000-0002-8288-9303; Lőrincz, András/0000-0002-1280-3447} } @article{MTMT:34045337, title = {Cross-Viewpoint Semantic Mapping: Integrating Human and Robot Perspectives for Improved 3D Semantic Reconstruction}, url = {https://m2.mtmt.hu/api/publication/34045337}, author = {Kopácsi, László and Baffy, Benjámin and Baranyi, Gábor and Skaf, Joul and Sörös, Gábor and Szeier, Szilvia and Lőrincz, András and Sonntag, Daniel}, doi = {10.3390/s23115126}, journal-iso = {SENSORS-BASEL}, journal = {SENSORS}, volume = {23}, unique-id = {34045337}, abstract = {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.}, year = {2023}, eissn = {1424-8220}, orcid-numbers = {Baffy, Benjámin/0000-0002-1775-180X; Lőrincz, András/0000-0002-1280-3447} } @article{MTMT:33947821, title = {Design of a Machine Learning System to Predict the Thickness of a Melanoma Lesion in a Non-Invasive Way from Dermoscopic Images}, url = {https://m2.mtmt.hu/api/publication/33947821}, author = {Szijártó, Ádám and Somfai, Ellák and Lőrincz, András}, doi = {10.4258/hir.2023.29.2.112}, journal-iso = {Healthcare Informatics Research}, journal = {Healthcare Informatics Research}, volume = {29}, unique-id = {33947821}, issn = {2093-369X}, year = {2023}, eissn = {2093-3681}, pages = {112-119}, orcid-numbers = {Somfai, Ellák/0000-0002-2218-8855; Lőrincz, András/0000-0002-1280-3447} } @article{MTMT:33416651, title = {Handling dataset dependence with model ensembles for skin lesion classification from dermoscopic and clinical images}, url = {https://m2.mtmt.hu/api/publication/33416651}, author = {Somfai, Ellák and Baffy, B. and Fenech, Kristian and Hosszú, R. and Korózs, D. and Pólik, Marcell and Sárdy, Miklós and Lőrincz, András}, doi = {10.1002/ima.22827}, journal-iso = {INT J IMAG SYST TECH}, journal = {INTERNATIONAL JOURNAL OF IMAGING SYSTEMS AND TECHNOLOGY}, volume = {33}, unique-id = {33416651}, issn = {0899-9457}, year = {2023}, eissn = {1098-1098}, pages = {556-571}, orcid-numbers = {Somfai, Ellák/0000-0002-2218-8855; Fenech, Kristian/0000-0002-8288-9303; Sárdy, Miklós/0000-0003-4306-5093; Lőrincz, András/0000-0002-1280-3447} } @{MTMT:33576741, title = {Temporal variables of speech in Parkinson’s Disease in three spontaneous speaking tasks}, url = {https://m2.mtmt.hu/api/publication/33576741}, author = {Bóna, Judit and Gosztolya, Gábor and Hoffmann, Ildikó and Klivényi, Péter and Tóth, Alinka and Svindt, Veronika and Tóth, László and Lőrincz, András}, booktitle = {Book of Abstracts : The 11th scientific conference with international participation Speech Research, Faculty of Humanities and Social Sciences, Zagreb, Croatia, December 8 - 10 2022}, unique-id = {33576741}, year = {2022}, pages = {28-29}, orcid-numbers = {Bóna, Judit/0000-0003-2369-1636; Gosztolya, Gábor/0000-0002-2864-6466; Klivényi, Péter/0000-0002-5389-3266; Svindt, Veronika/0000-0002-6027-9029; Tóth, László/0000-0003-0161-1375; Lőrincz, András/0000-0002-1280-3447} } @inproceedings{MTMT:33305314, title = {Structural Extensions of Basis Pursuit: Guarantees on Adversarial Robustness}, url = {https://m2.mtmt.hu/api/publication/33305314}, author = {Szeghy, David and Aslan, Mahmoud and Fóthi, Áron and Meszaros, Balazs and Milacski, Zoltán Ádám and Lőrincz, András}, booktitle = {DELTA: PROCEEDINGS OF THE 3RD INTERNATIONAL CONFERENCE ON DEEP LEARNING THEORY AND APPLICATIONS}, doi = {10.5220/0011138900003277}, unique-id = {33305314}, abstract = {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*.}, keywords = {Sparse coding; Stability theory; adversarial attack; Group Sparse Coding}, year = {2022}, pages = {77-85}, orcid-numbers = {Fóthi, Áron/0000-0002-1662-7583; Milacski, Zoltán Ádám/0000-0002-3135-2936; Lőrincz, András/0000-0002-1280-3447} } @misc{MTMT:33272087, title = {Perceived personality state estimation in dyadic and small group interaction with deep learning methods}, url = {https://m2.mtmt.hu/api/publication/33272087}, author = {Fenech, Kristian and Fodor, Ádám and Bergeron, Sean Paul and Saboundji, Rachid Rhyad and Oertel, Catharine and Lőrincz, András}, volume = {abs/2211.04979}, unique-id = {33272087}, year = {2022}, orcid-numbers = {Fenech, Kristian/0000-0002-8288-9303; Lőrincz, András/0000-0002-1280-3447} } @inproceedings{MTMT:33214815, title = {Towards creating a conversational memory for long-term meeting support: predicting memorable moments in multi-party conversations through eye-gaze}, url = {https://m2.mtmt.hu/api/publication/33214815}, author = {Tsfasman, Maria and Fenech, Kristian and Tarvirdians, Morita and Lőrincz, András and Jonker, Catholijn and Oertel, Catharine}, booktitle = {INTERNATIONAL CONFERENCE ON MULTIMODAL INTERACTION}, doi = {10.1145/3536221.3556613}, unique-id = {33214815}, year = {2022}, pages = {94-104}, orcid-numbers = {Tsfasman, Maria/0000-0001-5582-7636; Fenech, Kristian/0000-0002-8288-9303; Tarvirdians, Morita/0000-0003-4246-0016; Lőrincz, András/0000-0002-1280-3447; Jonker, Catholijn/0000-0003-4780-7461; Oertel, Catharine/0000-0002-8273-0132} } @misc{MTMT:33113731, title = {Opportunities and Limitations in the Analysis of Autistic Phenotypes - A Conceptual Review}, url = {https://m2.mtmt.hu/api/publication/33113731}, author = {Éva, Mariczáné Csanádi* and Fóthi, Ábel and Péter, Pollner and Lőrincz, András and Bálint, Kovács}, unique-id = {33113731}, year = {2022}, orcid-numbers = {Lőrincz, András/0000-0002-1280-3447} }