We invite high-quality papers covering the topics listed below.
Topics of interest include (but not limited to) applications of computer vision and machine learning to:
Feature extraction, classification and segmentation of materials image data;
Semantic data structuring: merging data and explicit knowledge for microstructure representations, including physics-informed machine learning for image analysis;
Establishing processing-structure-property relationships with extracted features or latent variables;
Spatiotemporal modeling for feature detection, tracking, and dynamic analysis in time-resolved imaging;
Data fusion from multimodal characterization, hyperspectral imaging, and/or a combination of simulated and experimental image data;
Creating synthetic microstructure images via generative models including VAE, GAN, and diffusion models;
Uncertainty quantification and calibration of classification and segmentation model predictions;
Explainable AI for microscopy data;
Representation learning, federated learning, self-supervised learning, or learning with rich labels such as text or quantitative material properties;
Efficient annotation methods and next-generation interactive model-driven microscope user interfaces;
Benchmark datasets.
Accepted papers will be presented as orals or posters at the workshop.
Important Dates
Paper Submission Due: March 22, 2024 (11:59pm PST) (closed)
Notification to Authors: April 3, 2024 (11:59pm PST) (notifications sent out)
Camera-ready paper due: April 11, 2024 (11:59pm PST) (submission instructions sent out)
Submission Guidelines
We invite two types of submissions:
Long archival paper: 4+ pages, should not exceed 8 pages, including figures and tables.
The long paper is for presenting novel ideas and mature work supported with experiments and analyses. A long paper will be published in the CVPR 2024 workshop proceedings.
Short non-archival paper: should not exceed 4 pages (excluding references).
The short paper is intended for sharing original early stages ideas, promising research or applications, pilot studies, work in progress, new datasets. Accepted short papers will not be included in the CVPR 2024 workshop proceedings.
Reviewing is double blind and there is no rebuttal period.
Authors are expected to remove their names and affiliations in the submitted version. Authors should also make a reasonable effort to anonymize the submitted code, data, and manuscript. If the paper is accepted, authors are expected to replace the submitted code/dataset with a non-anonymized version or link to a public GitHub repository.
Paper submissions must adhere to the CVPR 2024 guidelines regarding style, format, length restrictions (4 page limit for a short paper, 8 pages for a long paper), and dual submissions. By submitting a manuscript to the workshop, the authors acknowledge that it has not been previously published or accepted for publication in substantially similar form in any peer-reviewed venue including journal, conference or workshop.
Submission link: https://cmt3.research.microsoft.com/CV4MS2024
Upon acceptance, each paper must be registered under an author full, in-person registration type. One registration may cover multiple papers. Registration details can be found here.
Note that everyone must be registered in advance to attend the workshop (even speakers).