Content-Based Multimedia Indexing
June 28-30, 2021 - Lille, France

Important dates

Regular (full / short) paper submission 15 January, 2021 01 March, 2021
Special session paper submission 15 January, 2021 01 March, 2021
Demo paper submission 15 January, 2021 01 March, 2021
Notification of acceptance (all papers) 15 March, 2021
Camera-ready papers due (all papers) 30 March, 2021

Call for papers — Regular sessions

Authors are encouraged to submit previously unpublished research papers in the broad field of content-based multimedia indexing and applications. We wish to highlight significant contributions addressing the main problem of search and retrieval but also the related and equally important issues of multimedia content management, user interaction, large-scale search, learning in retrieval, social media indexing and retrieval. Additional special sessions are planned in areas such as deep learning for retrieval, social media retrieval, cultural heritage, surveillance and security.

Authors can submit full length (6 pages - to be presented as oral presentation) or short papers (4 pages - to be presented as posters). The submissions are peer reviewed in a single blind process. The language of the conference is English. The CBMI 2021 conference adheres to the IEEE paper formatting guidelines. When preparing your submission, please follow the IEEE guidelines given by IEEE at the Manuscript Templates for Conference Proceedings.

The CBMI proceedings are traditionally indexed and distributed by IEEE Xplore and ACM DL. In addition, authors of certain best papers of the conference will be invited to submit extended versions of their contributions to a special issue of a leading journal in the field (e.g. MTAP - Springer), and other best papers will be invited to submit extended versions of their contributions in a book (ISTE/WILEY publisher).

Topics of interest to the CBMI community include, but are not limited to, the following:

  • Audio and visual and multimedia indexing
  • Multimodal and cross-modal indexing
  • Deep learning for multimedia indexing
  • Visual content extraction
  • Audio (speech, music, etc) content extraction
  • Identification and tracking of semantic regions and events
  • Social media analysis
  • Metadata generation, coding and transformation
  • Multimedia information retrieval (image, audio, video, text)
  • Mobile media retrieval
  • Event-based media processing and retrieval
  • Affective/emotional interaction or interfaces for multimedia retrieval
  • Multimedia data mining and analytics
  • Multimedia recommendation
  • Large scale multimedia database management
  • Summarization, browsing and organization of multimedia content
  • Personalization and content adaptation
  • User interaction and relevance feedback
  • Multimedia interfaces, presentation and visualization tools
  • Evaluation and benchmarking of multimedia retrieval systems
  • Applications of multimedia retrieval, e.g., medicine, lifelogs, satellite imagery, video surveillance
  • Cultural heritage applications

Call for papers — Special sessions

Authors can submit papers (up to 6 pages) to be presented as oral presentation during special sessions. On the CBMI 2021 submission website, authors must specify the track corresponding to the special session to which they submit their paper. The submissions will be peer reviewed in a single blind process. The language of the conference is English. The CBMI 2021 conference adheres to the IEEE paper formatting guidelines. When preparing your submission, please follow the IEEE guidelines given by IEEE at the Manuscript Templates for Conference Proceedings.

Special sessions papers will be included in the CBMI proceedings like regular papers. The CBMI proceedings are traditionally indexed and distributed by IEEE Xplore and ACM DL. In addition, authors of certain best papers of the conference will be invited to submit extended versions of their contributions to a special issue of a leading journal in the field (e.g. MTAP - Springer), and other best papers will be invited to submit extended versions of their contributions in a book (ISTE/WILEY publisher).

Special session: Bio-inspired circuits, systems and algorithms for multimedia

CBMI aims at bringing together the various communities involved in all aspects of content-based multimedia indexing for retrieval, browsing, management, visualization and analytics. The Special Session on bio-inspired circuits, systems and algorithms will be a mini-venue, focusing on the state-of-the-art and research direction of the emerging field of bio-inspired circuits, systems and algorithms and in the envisioned applications and breakthrough that the application of this bioinspired technology and algorithms can bring to the content-based multimedia indexing field development.

Special session papers, which can be invited or submitted, will supplement the regular research papers and be included in the proceedings of CBMI 2021.

The special session would include four to five papers, which can be invited, or regular submissions. In order to ensure the high quality of all conference papers, all papers submitted to special session will be peer-reviewed through the standard review process, including invited papers. If the special session has many high-quality submissions, some of the submissions may potentially be moved to some regular sessions.

Authors are encouraged to submit previously unpublished research papers including bioinspired circuits, systems and algorithms that address the use of the bioinspired technology and algorithms in the areas of search and retrieval, multimedia content management, user interaction, large-scale search, learning in retrieval, social media indexing and retrieval, surveillance and security. Other applications areas of interest to the CBMI community not listed above maybe considered.

Organizer : Teresa Serrano Gotarredona, IMSE-CNM — terese@imse-cnm.csic.es

Special session: Content-based learning in astrophysics (CoBLA)

New observing and acquisition technologies provide a sharper and deeper view of the sky, with gigantic amounts of data,very high resolution sensors, very high frame rates... For instance, when in full operation, the Cherenkov Telescope Array (CTA)will produce a data volume of 210 PB of raw data per year, Legacy Survey of Space and Time (LSST) with acamera of about 3.2 gigapixelsor also Gravitational Waves physics (LIGO,Virgo, KAGRA)both requirereal time analysisof very large data streams.In this context, processing and physics model understanding needs are getting more and more crucial and this has encouraged the development of new data-based solutions that complement the usual paradigm of model-based data analysis. As a result, approaches widely used in the multimedia field, such as data mining, deep learning... have been transferred to the field of astrophysics providing interesting results to address problems such as the formation and the dynamics of galaxies, the enigmatic phenomena associated with black holes...This special session aims to bring together researchers working on analysis, indexing, and mining of datain the field of astrophysics, provides them a venue for sharing novel ideas and discuss their most recent works and promotes exchanges between computer scientists and astrophysicists. Topics of interest include (but are not limited to):

  • Supervised learning: classification and regression
  • Unsupervised learning: clustering and dimensionality reduction
  • Correlation between celestial events
  • Real time, on-site or on-board processing
  • Learning physical models from data
  • From simulated to actual data
  • Time-Series Analysis
  • Management of astrophysics data

Organizers :

  • Alexandre Benoit, Savoie Mont Blanc University, LISTIC, Annecy, France — alexandre.benoit@univ-smb.fr
  • Patrick Lambert, Savoie Mont Blanc University, LISTIC, Annecy, France — patrick.lambert@univ-smb.fr
  • Elena Cuoco, European Gravitational Observatory, Pisa, Italy — elena.cuoco@ego-gw.it

Special session: Mining and indexing multimedia data for remote sensing of the environment and our changing planet

This special session is dedicated to mining and indexing of complex data for monitoring the environment and our changing planet, in particular the use of hyperspectral satellite imagery for remote sensing applications. Several targeted applications including environment preservation, extreme events detection and forecasting, air quality measure, pollution detection and tracking are of interest. Of particular importance due to its social impact (high levels of air pollution lead to increase in public health issues) and policy enforcement implications, the monitoring of air quality (cleanliness of the air or pollution level) is also challenging in terms of data processing due to the complexity and vast volume of data available and the requirement of real time (or NRT) availability of products for application.

High performance mining and indexing of images for remote sensing applications could have significant implication for citizen and decision makers by contributing to fast prediction and reliable air quality monitoring systems. These should be able to cover the local and regional scale through intelligent extrapolation of sparse ground-based in situ measurement networks combined with large scale satellite observations and model simulations.

The objective of the special session is attracting papers in this very challenging context where image mining and indexing meet atmospheric physics and chemistry and remote sensing. The special session encourages papers that deal with the challenges of indexing and mining data for remote sensing. One of them is building new public datasets and benchmarks, which are hot priority of hyperspectral remote sensing community. Another one is developing indexing and mining hyperspectral image classification with big data and without or with very limited ground truth. Hyperspectral image datasets are gathered from satellite observations and interpreted thanks to meteorological model. The ground truth, generally very limited when available, are particulate matter concentration in the atmospheric boundary layer, from ground-based observation networks. These interdependent challenges will contribute to improve air quality modelling and open pathways to efficient experimental environment for our approaches and for the approaches of the international remote sensing community. We encourage papers in the domain of machine learning (ML) techniques, and more particularly deep learning. ML will operate on data (space based, model, ground network) in order to provide a fast prediction, statistically optimized model of air quality able to provide relevant information to decision makers at the local and regional scales.

Another challenge the special session would like to investigate is hybrid approach combining statistical optimization based on deep learning (Artificial Intelligence) and model scaling to infer Particulate Matter near the surface from satellite observations performed by different sensors.

Topics of interest include (but are not limited to):

  • Hyperspectral images for remote sensing
  • Big data in the context of changing planet
  • Fast prediction model
  • Local and regional air quality assessment
  • Detection and tracking of Particulate Matter
  • Benchmaks and evaluation protocols
  • Machine learning (e.g. deep learning) on atmospheric images
  • Satellite data, including hyperspectral images
  • Correlation of hybrid data from sensors
  • Applications : air quality, aerosol, ...

Organizers :

  • Pr. Jérôme RIEDI, ICARE Data and Services Center, University of Lille, France — jerome.riedi@univ-lille.fr
  • Dr. Suzanne CRUMEYROLLE, Laboratoire d’Optique Atmosphérique, University of Lille, France — suzanne.crumeyrolle@univ-lille.fr
  • Dr. Robert C. LEVY, Climate and Radiation Laboratory, NASA Goddard Space Flight Center, USA
  • Dr. Otto HASEKAMP, Earth Science Groupe, SRON, Netherlands
  • Dr. Shan ZENG, Science Systems and Applications, Inc., USA
  • Dr Vincent Oria, New Jersey Institute of Technology, USA — vincent.oria@njit.edu

Call for demos

Authors are encouraged to submit previously unpublished demonstration papers in the broad field of content-based multimedia indexing and applications. Demonstration pages should be up to 4 pages in length and highlight interesting and novel demos of CBMI-related technologies. CBMI 2021 especially seeks novel and compelling demonstrations that highlight the potential positive impact of content-based multimedia indexing in daily life. Demonstration papers will be reviewed according to criteria such as novelty, interestingness, applications of or enhancements to state-of-the-art, and potential impact. The submissions are peer reviewed in a single blind process. The language of the demonstration is English. CBMI 2021 conference adheres to the IEEE paper formatting guidelines. When preparing your submission, please follow the IEEE guidelines given by IEEE at the Manuscript Templates for Conference Proceedings.

CBMI proceedings are traditionally indexed and distributed by IEEE Xplore and ACM DL. In addition, authors of the best papers of the conference will be invited to submit extended versions of their contributions to a special issue of a leading journal in the field.

Topics of interest to the CBMI community include, but are not limited to, the following:

  • Audio and visual and multimedia indexing
  • Multimodal and cross-modal indexing
  • Deep learning for multimedia indexing
  • Visual content extraction
  • Audio (speech, music, etc) content extraction
  • Identification and tracking of semantic regions and events
  • Social media analysis
  • Metadata generation, coding and transformation
  • Multimedia information retrieval (image, audio, video, text)
  • Mobile media retrieval
  • Event-based media processing and retrieval
  • Affective/emotional interaction or interfaces for multimedia retrieval
  • Multimedia data mining and analytics
  • Multimedia recommendation
  • Large scale multimedia database management
  • Summarization, browsing and organization of multimedia content
  • Personalization and content adaptation
  • User interaction and relevance feedback
  • Multimedia interfaces, presentation and visualization tools
  • Evaluation and benchmarking of multimedia retrieval systems
  • Applications of multimedia retrieval, e.g., medicine, lifelogs, satellite imagery, video surveillance
  • Cultural heritage applications

Call for sponsors

Available soon.

Student Participation

We strongly encourage students to participate in CBMI-21 event and submit their research. We strongly believe in their power, and they are the future of the research in content-based multimedia indexing. For these reasons and for the first time in the history of the conference, the 2021 edition of CBMI with support of ACM SIGMM (www.sigmm.org), will sponsor several students, authors of papers submitted and accepted by CBMI-21, and being corresponding author of a paper. The student status will be recognized only to PhDs and master students. Certain students will be totally sponsored, including registration fees, accommodations and travel expenses. Other students will be partly sponsored, including substantial reduction in the registration fees, no cut-backs to the students’ conference experience, and budget accommodation options and arrangement for room sharing.

Furthermore, in this edition, prizes will be awarded to the best student paper presentations of CBMI2021. All the participants registered to CBMI2021 as student will be automatically admitted to the selection for the awards. All the papers accepted as presentation by the Technical Program Committee (TPC), from students will be considered for the awards, given that a full-paper manuscript has been submitted.

The student paper awards committee will evaluate the nominated contributions. The evaluation criteria will be independently established by the Committee before the conference and will take into account the scientific content as well as the technical quality of the papers.

The Student Paper Awards will be announced during the Gala Dinner, on June 29th.