HMM-QoE 2023
An IEEE ICASSP Workshop on
Humans, Machines and Multimedia - Quality of Experience and Beyond
June 5, 2023, Jupiter Ballroom
Workshop Scope
Since humans are typically the end users of multimedia systems, in the last ~20 years, an important focus of the research activity on multimedia systems has been how to measure the quality as perceived by humans via subjective tests and objective measurements. Objective quality metrics have been developed to estimate the quality as perceived by humans and the design of the whole transmission chain is based on the maximization of such metrics. Several open issues still exist in this domain, such as accurately estimating the perceived quality based on content features and QoS measurements in the media delivery chain.
Moreover, in recent years, the advent of machine-to-machine communications and machine learning opened new scenarios where “machines” are the end user of multimedia data. Machines process information differently from humans, for instance being able to find patterns in unstructured data, but also being easily “fooled” by well-engineered data perturbations. Relevant scenarios include autonomous driving, surveillance networks, medical diagnostic systems and robotic surgery. In some cases, the content is used by both humans and machines. Hence, suitable metrics should be designed to cover both cases and system optimisation should aim at meeting the joint requirements. For instance, the efficiency of video compression strategies, currently only aiming at satisfying human based quality metrics, should be improved for this scenario taking into account the purpose of machine vision systems.
This workshop tackles the challenge of video quality assessment and provision for both humans and machines, highlighting the common aspects and the orthogonal challenges.
This workshop will solicit contributions in the following areas:
Multimedia quality measurement and modeling for joint human and machine use
Quality metrics for multimedia based automatic decision systems.
Neural network misclassification rate vs. compression rate
Quality assessment for visual data representations non interpretable by humans
Video compression for humans and machines
Eliciting quality requirements for humans and machines in specific use cases
Video transmission for humans and ML based systems
Measuring quality in multimedia machine-to-machine communication systems.
Datasets for quality assessment for humans and machines
Contributions sharing also relevant datasets or source code are also welcome.
Workshop Organizers
Program Committee Members
Luigi Atzori, University of Cagliari, Italy
Lina Karam, Arizona State University, USA
Patrick Le Callet, University of Nantes, France
Mikołaj Leszczuk, AGH University of Science and Technology, Krakow, Poland
Silvia Rossi, CWI, the Netherlands
Alan Guedes, University College London, UK
Nabeel Khan, University of Chester, UK
Pedro Gomes, University College London, UK
Program
Welcome and Opening (8:30am-8:35am, Maria Martini)
Keynote (8:35am-9:20am)
Visual Coding for Humans and Machines, Ivan Bajić (Simon Fraser University) - Slides
Abstract: Visual content is increasingly being used for more than human viewing. For example, traffic video is automatically analyzed to count vehicles, detect traffic violations, estimate traffic intensity, and recognize license plates; images uploaded to social media are automatically analyzed to detect and recognize people, organize images into thematic collections, and so on; visual sensors on autonomous vehicles analyze captured signals to help the vehicle navigate, avoid obstacles, collisions, and optimize their movement. The above applications require continuous machine-based analysis of visual signals, with only occasional human viewing, which necessitates rethinking the traditional approaches for image and video compression. This talk is about coding visual information in ways that enable efficient usage by machine learning models, in addition to human viewing. We will touch upon recent rate-distortion results in this field, describe several designs for human-machine image and video coding, and briefly review related standardization efforts.
Biography: Ivan V. Bajić is a Professor of Engineering Science and co-director of the Multimedia Lab at Simon Fraser University, Canada. His research interests include signal processing and machine learning with applications to multimedia processing, compression, and collaborative intelligence. His group’s work has received several research awards, including the 2023 TCSVT Best Paper Award, conference paper awards at ICME 2012, ICIP 2019, and MMSP 2022, and other recognitions (e.g., paper award finalist, top n%) at Asilomar, ICIP, ICME, ISBI, and CVPR. Ivan is currently the Chair of the IEEE Multimedia Signal Processing Technical Committee. He has served on the organizing and/or program committees of the main conferences in his field, and has received several awards in these roles, including Outstanding Reviewer Award (six times), Outstanding Area Chair Award, and Outstanding Service Award. He was on the editorial boards of the IEEE Transactions on Multimedia and IEEE Signal Processing Magazine, and is currently a Senior Area Editor of the IEEE Signal Processing Letters.
Oral Session Part I (9:20am-10am, Chair: Angelo Coluccia)
LCCM-VC: Learned Conditional Coding Modes for Video Compression, Hadi Hadizadeh (Simon Fraser University); Ivan Bajić (Simon Fraser University)
Predicting CNN learning accuracy using chaos measurement, Rémi Piau (INRIA); Thomas Maugey (INRIA); Aline Roumy (INRIA)
Coffee Break
Oral Session Part II (10:20am-11:20am, Chair: Pier Luigi Dragotti)
A study on the impact of virtual reality on user attention, Sara Baldoni (University of Padova); Mahmoud Z. A. Wahba (University of Padova); Marco Carli (Università degli Studi Roma Tre); Federica Battisti (University of Padova)
Quality of Things based machine learning for the M-IoT Applications, Shaymaa Al-Juboori (University of Plymouth); Ali Al-nuaimi (University of Plymouth); Amulya Karaadi (University of Plymouth); Is-Haka Mkwawa (University of Plymouth); Jianwu Zhang (Hangzhou Dianzi University); Lingfen Sun (University of Plymouth)
A Clustered Federated Learning Approach for Estimating the Quality of Experience of Web Users, Simone Porcu (University of Cagliari); Alessandro Floris (University of Cagliari); Luigi Atzori (University of Cagliari)
Invited Talks (11:20am-12pm, Chair: Maria Martini)
Image and Video Coding for Machines, Kristian Fischer (Friedrich-Alexander-Universität Erlangen-Nürnberg) - Slides
Opportunity in Difficulty: Post Pandemic Practices for QoE Testing, Patrick Le Callet (Nantes Université, LS2N)