Invited Talks















20210325_173827.mp4

2021


October

Autumn School MALVIC: Machine Learning and Vision for Industrial Applications

Marius Leordeanu was a keynote speaker, with the talk titled Towards Unsupervised Learning in Space and Time: From Metric Depth Estimation and Semantic Segmentation to Complete 4D Scene Understanding.

The goal of this Autumn School MALVIC is to bring together pioneering international scientists in machine learning and computer vision with both academia and practitioners from the industrial fields on a unique setting for the discussion and demonstration of practical, hands-on machine learning and vision research and development. Offshore industrial applications and industrial process scenarios are examples for the autumn school target.

Slides

July

ML@AIMAS Summer School 2021

Marius Leordeanu, the scientific coordinator of our project, alongside PhD students Alina Marcu and Dragos Costea hold the Computer Vision theoretical and practical sessions.

Summer School on Image Processing 2021

Radu Ionescu, a member of our partner, the University of Bucharest, presented the "Anomaly detection in video: approaches and challenges" talk, discussing several state-of-art approach for video anomaly detection and future challenges.

March

ESA (European Space Agency) - workshop: European Space Agency Workshop on Artificial Intelligence

Marius Leordeanu was a keynote speaker, with the talk titled Learning to See the World from Above: From Semantic Interpretation of Satellite Images to Unsupervised Scene Understanding from Flight

Slides




2020

December

Romanian AI Days is an annual event dedicated to the AI community in Romania. Its aim is to strengthen the Romanian AI community and increase its impact in the international AI ecosystem, by bringing together researchers, entrepreneurs and policy makers from Romania and abroad.

Marius Leordeanu, the scientific coordinator of our project, was invited as a keynote speaker at this wonderful event and he presented our latest research results on unsupervised learning for multi-task scene understanding in the context of intelligent drones, with focus on our recent AAAI 2021 and ACCV 2020 papers.

The talk can also be seen on YouTube (video on the right), in a slightly different version, given immediately after Romanian AI Days, at an online event organized by Under Construction, one of the top Robotics clubs at the National level, formed by high-school students from Carabela, Targoviste - one of the best High-schools in Romania.



2019

September

SSIMA Re:Imagine Healthcare

SSIMA is the annual international MedTech festival in Eastern Europe, covering topics related to medical technology, such as: Deep Learning and Big Data in Healthcare, Medical Imaging, Medical Sensors and Robots and AI-based Data Analysis. It covers these topics from academic, business and public policy perspectives, with the explicit aim of increasing relationships between Eastern European academics, companies and governments and their global counterparts.

Marius Leordeanu was invited as a keynote speaker at SSIMA 2019, where he presented Towards Unsupervised Learning in Space and Time: A Computer Vision Approach, whilst Alina Marcu and Dragos Costea hold the practical sessions Tensorflow-Keras programming course.


July

EEML 2019

Marius Leordeanu was invited as a keynote speaker at EEML 2019 where he talked about Unsupervised Learning for Computer Vision, at the University Politehnica of Bucharest.

Photo taken by Dr. Rahul Sukthankar, Director of Google AI Perception Research.

Slides

January

Marius Leordeanu was invited as a keynote speaker at Dagstuhl Seminar on Joint Processing of Language and Visual Data for Better Automated Understanding, Seminar id: 19021, Dagstuhl School, Germany, where he presented our latest research titled: Mining for meaning: from vision to language through multiple networks consensus.