Call for Papers

The goal of this workshop is to bring together a broad and representative group of computer vision researchers with an interest in non-Euclidean geometry for deep learning to elicit lively and impactful discussions on a wide range of topics, including: 

• Algorithms based on hyperbolic and hyperspherical geometries and their performance analysis, including novel optimization techniques. 

• Comparisons between different techniques and methods for hyperbolic and hyperspherical geometries (and their Euclidean counterparts)

• Proposals of taxonomies and terminologies for hyperbolic and hyperspherical geometry in computer vision, with discussion of annotations and metrics.

•  Novel datasets for performance evaluation and/or empirical analyses of existing methods.

• Supervised and unsupervised methods, as well as techniques for semi-supervised and self-supervised learning applied to non-Euclidean geometries.

• Applications of non-Euclidean geometry to topics such as event recognition, activity detection, object recognition, and tracking, among others.

2 No Proceedings track: extended abstracts or published work

Extended abstract submissions should be in the ECCV format and are limited to 4 pages, including figures and tables. Additional pages containing only cited references are allowed. Please download the ECCV 2024 Author Kit for detailed formatting instructions. This track welcomes submissions beyond traditional research papers, including early-stage results, insightful negative findings, opinion pieces, and novel datasets. Moreover, we welcome papers published at peer-reviewed venues and cover topics related to the workshop. ECCV 2024 paper style is not required. Works submitted to this track will not be included in the ECCV proceedings, providing an opportunity to share preliminary work or important milestones to foster discussion in the research community.

Review Process

Reviews are double-blind and conducted through CMT. All submissions will receive a minimum of two reviews.