TY - CHAP AU - Sulyok, András Attila AU - Karacs, Kristóf TI - Towards Using Fully Observable Policies for POMDPs T2 - 2022 2nd International Conference on Computing and Machine Intelligence (ICMI) PB - IEEE SN - 9781665474832 PY - 2022 SP - 1 EP - 5 PG - 5 DO - 10.1109/ICMI55296.2022.9873768 UR - https://m2.mtmt.hu/api/publication/33097232 ID - 33097232 LA - English DB - MTMT ER - TY - JOUR AU - Németh, János Tibor AU - Nyitrai, Beatrix AU - Karacs, Kristóf AU - Szabó, Dorottya AU - Ecsedy, Mónika AU - Szalai, Irén AU - Tóth, Gábor AU - Sándor, Gábor László AU - Magyar, Márton AU - Benyó, Fruzsina AU - Papp, András TI - OCT-leletek telemedicinális értékelésének pontossága cukorbetegekben JF - SZEMÉSZET J2 - SZEMÉSZET VL - 159 PY - 2022 IS - 2 SP - 64 EP - 68 PG - 5 SN - 0039-8101 DO - 10.55342/szemhungarica.2022.159.2.64 UR - https://m2.mtmt.hu/api/publication/33039528 ID - 33039528 LA - Hungarian DB - MTMT ER - TY - JOUR AU - Gelencsér-Horváth, Anna AU - Kopácsi, László AU - Varga, Viktor AU - Keller, Dávid AU - Dobolyi, Árpád AU - Karacs, Kristóf AU - Lőrincz, András TI - Tracking Highly Similar Rat Instances under Heavy Occlusions: An Unsupervised Deep Generative Pipeline JF - JOURNAL OF IMAGING J2 - J IMAGING VL - 8 PY - 2022 IS - 4 PG - 20 SN - 2313-433X DO - 10.3390/jimaging8040109 UR - https://m2.mtmt.hu/api/publication/32783352 ID - 32783352 N1 - Faculty of Information Technology and Bionics, Pázmány Péter Catholic University, Práter utca 50/A, Budapest, 1083, Hungary Department of Artificial Intelligence, Faculty of Informatics, Eötvös Loránd University, Pázmány Péter Sétány 1/C, Budapest, 1117, Hungary Laboratory of Neuromorphology, Department of Anatomy, Histology and Embryology, Semmelweis University, Budapest, 1094, Hungary ELKH-ELTE Laboratory of Molecular and Systems Neurobiology, Eötvös Loránd Research Network, Eötvös Loránd University, Brussels, 1000, Belgium Department of Physiology and Neurobiology, Eötvös Loránd University, Pázmány Péter Sétány 1/A, Budapest, 1117, Hungary Cited By :1 Export Date: 20 February 2024 Correspondence Address: Gelencsér-Horváth, A.; Faculty of Information Technology and Bionics, Práter utca 50/A, Hungary AB - Identity tracking and instance segmentation are crucial in several areas of biological research. Behavior analysis of individuals in groups of similar animals is a task that emerges frequently in agriculture or pharmaceutical studies, among others. Automated annotation of many hours of surveillance videos can facilitate a large number of biological studies/experiments, which otherwise would not be feasible. Solutions based on machine learning generally perform well in tracking and instance segmentation; however, in the case of identical, unmarked instances (e.g., white rats or mice), even state-of-the-art approaches can frequently fail. We propose a pipeline of deep generative models for identity tracking and instance segmentation of highly similar instances, which, in contrast to most region-based approaches, exploits edge information and consequently helps to resolve ambiguity in heavily occluded cases. Our method is trained by synthetic data generation techniques, not requiring prior human annotation. We show that our approach greatly outperforms other state-of-the-art unsupervised methods in identity tracking and instance segmentation of unmarked rats in real-world laboratory video recordings. LA - English DB - MTMT ER - TY - JOUR AU - Radványi, Mihály Gergely AU - Karacs, Kristóf TI - Peeling off image layers on topographic architectures JF - PATTERN RECOGNITION LETTERS J2 - PATTERN RECOGN LETT VL - 135 PY - 2020 SP - 50 EP - 56 PG - 7 SN - 0167-8655 DO - 10.1016/j.patrec.2020.04.023 UR - https://m2.mtmt.hu/api/publication/31306422 ID - 31306422 N1 - Funding Agency and Grant Number: European Union - European Social Fund [EFOP-3.6.3VEKOP-16-2017-00 0 02, 2018-1.2.1-NKP-00008]; PPCU - University of National Excellence ITK Research Faculty Funding text: This research has been partially supported by the European Union, co-financed by the European Social Fund (EFOP-3.6.3VEKOP-16-2017-00 0 02 and 2018-1.2.1-NKP-00008). The authors also acknowledge PPCU - University of National Excellence ITK Research Faculty. LA - English DB - MTMT ER - TY - JOUR AU - Németh, János Tibor AU - Maka, Erika AU - Szabó, Dorottya AU - Somogyvári, Zsolt AU - Kovács, Gábor AU - Tóth, Gábor AU - Papp, András AU - Karacs, Kristóf AU - Nagy, Zoltán Zsolt TI - Működő telemedicinális szemészeti szűrőprogramok és lehetőségek hazánkban JF - IME J2 - IME VL - 18 PY - 2019 IS - 8 SP - 46 EP - 51 PG - 6 SN - 1588-6387 UR - https://m2.mtmt.hu/api/publication/30954730 ID - 30954730 LA - Hungarian DB - MTMT ER - TY - GEN AU - Németh, János Tibor AU - Maka, Erika AU - Szabó, Dorottya AU - Somogyvári, Zsolt AU - Kovács, Gábor AU - Papp, András AU - Karacs, Kristóf AU - Nagy, Zoltán Zsolt TI - Telemedicina a szemészetben, múlt, jelen és jövő PY - 2019 UR - https://m2.mtmt.hu/api/publication/30926541 ID - 30926541 LA - Hungarian DB - MTMT ER - TY - JOUR AU - Stubendek, Attila AU - Karacs, Kristóf TI - Shape Recognition Based on Projected Edges and Global Statistical Features JF - MATHEMATICAL PROBLEMS IN ENGINEERING J2 - MATH PROBL ENG VL - 2018 PY - 2018 PG - 18 SN - 1024-123X DO - 10.1155/2018/4763050 UR - https://m2.mtmt.hu/api/publication/3383887 ID - 3383887 LA - English DB - MTMT ER - TY - CHAP AU - Radványi, Mihály Gergely AU - Karacs, Kristóf ED - Tetzlaff, R TI - Saliency based Attention Mechanism for Topographic Architectures T2 - 15th International Workshop on Cellular Nanoscale Networks and their Applications PB - VDE Verlag CY - Berlin SN - 3800742527 PY - 2016 UR - https://m2.mtmt.hu/api/publication/3335502 ID - 3335502 LA - English DB - MTMT ER - TY - CHAP AU - Karacs, Kristóf AU - Stubendek, Attila AU - Radványi, Mihály Gergely TI - An Integrated Assistance Tool for Visual Impairment T2 - Proceedings of the Workshop on Information Technology and Bionics PB - Pázmány University ePress CY - Budapest SN - 9789638988034 PY - 2015 SP - 87 EP - 90 PG - 4 UR - https://m2.mtmt.hu/api/publication/2964429 ID - 2964429 LA - English DB - MTMT ER - TY - CHAP AU - Karacs, Kristóf AU - Radványi, Mihály Gergely AU - Stubendek, Attila AU - Bezanyi, B ED - EMB, null ED - IEEE, Circuits Systems Society TI - Learning hierarchical spatial semantics for visual orientation devices T2 - 10th IEEE Biomedical Circuits and Systems Conference, BioCAS 2014 PB - IEEE CY - Piscataway (NJ) SN - 9781479923465 T3 - IEEE 2014 Biomedical Circuits and Systems Conference, BioCAS 2014 - Proceedings PY - 2014 SP - 141 EP - 144 PG - 4 DO - 10.1109/BioCAS.2014.6981665 UR - https://m2.mtmt.hu/api/publication/3029261 ID - 3029261 AB - Complexity of understanding a visual scene is the single biggest challenge in creating intelligent devices for visually impaired people. The requirement of real time operation makes it inevitable to design algorithms that obey the computing and memory limits of available hardware. We present a hierarchical scene understanding system implemented on a vision system chip. It is restricted to extract specific information for predefined categories of visual scenes, but it is general enough to be able to learn quickly and autonomously. Patches having potential discriminative information are extracted using a hierarchical peeling method. Object groups are created based on proximity and size of the patches. Objects are classified using different classifiers and the votes are combined using a mixture of experts network. Experimental validation has been carried out on authentic image flows recorded by blind subjects. © 2014 IEEE. LA - English DB - MTMT ER -