Objava u časopisu Sensors

Znanstveni članak “Active Player Detection in Handball Scenes Based on Activity Measures” autora M. Pobar, M. Ivašić-Kos objavljen je u otvorenom pristupu u časopisu MDPI Sensors, u specijalnom izdanju “Advanced Computational Intelligence for Object Detection, Feature Extraction and Recognition in Smart Sensor Environments“, Sensors 202020(5), 1475 i dostupan je na webu:

Sažetak: https://www.mdpi.com/1424-8220/20/5/1475
PDF: https://www.mdpi.com/1424-8220/20/5/1475/pdf

Intervju za portal netokracija.com

Na portalu netokracija.com objavljen je intervju s M. Ivašić-Kos pod naslovom “Data science u sportu: prof. dr. sc. Marina Ivašić-Kos analizira rukometne akcije uz strojno učenje” u kojem su prezentirani rezultati projekta RAASS i dan plan budućih istraživanja. prof. Ivašić-Kos je dala i osvrt vezano za razvoj Data scienca i budućnost dubokog učenja. Članak se nalazi na linku: https://www.netokracija.com/strojno-ucenje-rukomet-wids-164836

Predstavljanje rada na meðunarodnom skupu ICPRAM 2020, Valletta

Na međunarodnoj konferneciji ICPRAM 2020 – 9th International Conference on Pattern Recognition Applications and Methods, 2020, održanoj 22.-24. veljače u Valletti, Malta, sudjelovale su članice tima Kristina Host i Marina Ivašić-Kos. Na konferenciji su predstavile rad “Tracking handball players with the DeepSORT algorithm“. Marina Ivašić-Kos je bila i voditeljica sekcije Data Mining and Algorithms for Big Data.

Predstavljanje radova na znanstvenom skupu ICT Innovations Conference 2019

Na znanstvenom skupu ICT Innovations Conference 2019, održanom u Ohridu 17.-19. listopada, sudjelovali su voditeljica projekta Marina Ivašić-Kos te suradnik Saša Sambolek. Na skupu su predstavili radove “Object Detetction Using Synthesized Data”, čiji su autori Matija Burić, Goran Paulin i Marina Ivašić-Kos, te rad “Detection of Toy Soldiers Taken from a Bird’s Perspective Using Convolutional Neural Networks “, autora Saše Samboleka i Marine Ivašić-Kos.

Research class predavanja u sklopu HRZZ RAASS projekta

U četvrtak 26.09.2019. godine i petak 27.09.2019. godine na Odjelu za informatiku Sveučilišta u Rijeci održano je više Research class predavanja u sklopu HRZZ RAASS projekta.

Partner na projektu, gost pedavač prof. Jordi Gonzàlez, CVC, Barcelona predstaviti će svoja istraživanja povezana s temom projekta, a doktorski studenti Odjela za informatiku i suradnici na projektu RAASS predstavit će trenutne istraživačke aktivnosti i dobivene rezultate.

Raspored održanih predavanja:

Četvrtak, 26.09.2019

10:00 – 11:00  Jordi Gonzàlez, CVC, Barcelona (gost predavač): “Going beyond Deep Learning in Understanding Human Behaviors in Image Sequences”

12:00 – 13:30  Aktivnosti u sklopu RAASS projekta: Automatsko raspoznavanje akcija u sportskim scenama

  • Matija Burić:  Object Detection and Tracking in Handball Scenes
  • Goran Paulin:  Synthesized Data Generation
  • Matija Burić: Object Detection Using Synthesized Data
  • Saša Sambolek: Detection of toy soldiers taken from a bird’s perspective

 

 Petak, 27.09.2019

10:00 – 11:00 Jordi Gonzàlez iz CVC, Barcelona (gost predavač): “Towards a Visual Inference of Personality Traits based on Images Shared in Social Networks”

12:00 – 12:30 Sobodan Beliga: Multilingual Keyword Extraction

 

Pozvano predavanje u sklopu HRZZ RAASS projekta

U petak, 27.9.2019., na Odjelu za informatiku Sveučilišta u Rijeci,  gost pedavač prof. Jordi Gonzàlez, CVC, Barcelona i partner na projektu HRZZ RAASS projekta održao je pozvano predavanje na kojem je predstavio svoja istraživanja povezana s temom projekta.

Predavač: prof. Jordi Gonzàlez, Computer Vision Center, Univ. Autònoma de Barcelona, Bellaterra, Barcelona, Spain

Naziv predavanja: Towards a Visual Inference of Personality Traits based on Images Shared in Social Networks

Abstract: The social media, as a major platform for communication and information exchange, provides a rich repository of people’s opinions and sentiments about a vast spectrum of topics. Such knowledge is embedded in multiple facets, such as comments, tags, as well as shared image and video content. The analysis of such information either in the area of opinion mining, affective computing or sentiment analysis is playing an important role in computational social sciences, which aims to understand and predict human decision making and enables applications such as brand monitoring, market prediction, and even voting forecasts. However, the massive growth of photo and video sharing is increasingly eclipsing text on the leading visual social platforms. So visual communication is complementing and even supplanting the written word, and this visual language is a powerful way for people to express themselves. This talk will exploit the most recent image understanding models based on neural networks to process the vast amount of data generated by social users. These improvements will enable to know more accurately the social user’s demands and cultural-driven interests, eventually reaching some degree of personality trait description, depression detection and early suicidal tendencies estimation.