More than 60 researchers joined hybrid international conference on Multimedia Analysis and Pattern Recognition at USTH

Sunday - 17/10/2021 00:09

On October 15 and 16, 2021, the 4th International Conference on Multimedia Analysis and Pattern Recognition (MAPR 2021) took place successfully at USTH in a hybrid format with the participation of more than 60 researchers and students in the ICT field from Vietnam and others countries. 

The International Conference on Multimedia Analysis and Pattern Recognition (MAPR) is an annual event held since 2018, bringing together researchers and practitioners from both academia and industry around the world as well as in Vietnam for sharing their latest research findings, experimental results and consolidating potential collaborations on pattern recognition, multimedia analysis, and related areas. It also provides a unique opportunity to PhD students and students to interact with renowned and experienced researchers in the field, to discuss with them their ongoing research and career plans. 

In previous years, the conference was organized at the University of Information Technology, Vietnam National University-Ho Chi Minh City (for the 1st & 2nd MAPR) and at the Hanoi University of Science and Technology (for the 3rd one). This year, USTH is honored to be the host institution of the 4th Conference, taking place on October 15 and 16, 2021. Due to the Covid-19 pandemic, the conference was held in a hybrid format in which only a small group of participants were able to be present at the venue while the other researchers joined virtually via the Zoom platform.

Attending this event, there were Prof. Jean-Marc Lavest, Principal Rector of USTH; Assoc. Prof. Tran Dinh Phong, Co-Director of Research, Innovation and Technology Transfer of USTH; Assoc. Prof. Nguyen Duc Dung, Co-Director of ICT Department of USTH, Program Committee Co-chair. There were also representatives of co-organizers from Vietnamese Association on Pattern Recognition (VAPR), the University of Information Technology (VNU-HCM), Institute of Information Technology (VAST), International Research Institute MICA (Hanoi University of Science and Technology) and more than 60 Vietnamese and foreign researchers and students in the field of ICT.  

Speaking at the opening of the conference, Prof. Jean-Marc Lavest, Principal Rector of USTH, expressed his delight that USTH had an honor of being the host institution of the MAPR 2021. He said that “The subject of pattern recognition has animated the scientific community for many years; the speed of the evolution of environments, whether in terms of content, or even of power and speed of calculation, pushes the scientific investigation research further every day.” He emphasized the spirit of the conference, which is based on “exchanges, better knowledge between stakeholders and building networks that tomorrow will allow us together to make ambitious responses to large calls for international projects”.

DSC0062 2048
The opening speech of the MAPR 2021 by Prof. Jean-Marc Lavest, Principal Rector of USTH

According to Assoc. Prof. Nguyen Duc Dung, Co-Director of ICT Department of USTH, Program Committee Co-chair, due to the effect of the Covid-19 pandemic, the number of submissions for the MAPR 2021 is not as high as expected but the acceptance rate has increased, that shows a significant increase in the quality of the submissions. There were 33 submitted papers from researchers and students in Vietnam and other countries, of which the organization committee selected 24 papers with the content revolving around keywords such as deep learning, object detection methods, mathematical formula detection, hand action recognition…
DSC0085 2048

Presentation of the conference summary by Prof. Nguyen Duc Dung, Program Committee Co-chair

The highlight of the MAPR 2021 was the sharing sessions of two keynote speakers: Professor Inseop Na from National Program of Excellence in Software Center and Major of AI-Convergence (Korea) on the topic “Introduction of emotion datasets and trends in recognition research in order to think machine by themselves” and Professor Antoine Doucet, from L3i laboratory of the University of La Rochelle (France) with a talk entitled “Robust and multilingual analysis of digitized documents - a use case with historical newspapers”. In his presentation, Prof. Inseop Na gave an overview of the current state of emotion datasets and related technologies in the field of emotion recognition. He also suggested the kind of research needed for machines to have emotions on their own in the future.

Regarding the presentation of Professor Antoine Doucet, he presented recent advances in AI and automatic language processing enabling the textual content of digitized documents to be analyzed in a way that is robust to digitization errors. He took an example with the H2020 NewsEye project in order to show how researchers were able to create state-of-the-art results for the cross-lingual recognition and disambiguation of named entities (names of people, places, and organizations) in large corpora of historical newspapers written in 4 languages, written between 1850 and 1950.

c

Presentation of Prof. Antoine Doucet

Besides, at the conference, the researchers of 24 accepted papers also made presentations on their study findings. Many questions were addressed to the presenters, providing an opportunity for open discussion among the participants on relevant topics.

After two days of sharing and discussion, the conference of MAPR 2021 ended with the award ceremony for the best paper and the best student paper.

DSC0094 2048

Representatives of the Organizing Committee take souvenirs photo with participants via Zoom