Call for Posters
Deadline for Submission: May 24, 2017
Notification of Acceptance: June 12, 2017
Workshop date: July 21, 2017
The purpose of this workshop is to bring together researchers to explore visual recognition across the continuum between basic level categorization (object recognition) and identification of individuals (face recognition, biometrics) within a category population.
Topics of interest include the following:
- Novel datasets and data collection strategies for fine-grained categorization
- Appropriate error metrics for fine-grained categorization
- Constructing field-guides for visual categories
- Embedding human experts’ knowledge into computational models
- Fine-grained categorization with humans in the loop
- Transfer-learning from known to novel subcategories. Zero/one-shot recognition
- Domain-specific techniques that generalize to various other domains
- Attribute-based techniques for fine-grained categorization
- Using taxonomies to improve fine-grained categorization
- Part-sharing models for categorization/recognition
- Unsupervised subcategory discovery
- Learning of discriminative features for fine-grained categorization
- Multimodal (e.g. combined audio/video) techniques for fine-grained categorization
- Fine-grained categorization systems deployed in practice
Submission and Reviews:
We invite submission of 1.5-2 page extended abstracts (using the CVPR 2017 format) describing work in the domains suggested above or in closely-related areas. Accepted submissions will be presented as posters at the workshop. Authors may submit a draft of their poster as an optional third page. The extended abstract and poster sketch should be submitted as a single PDF file with no more than 3 pages. Reviewing of abstract submissions will be double-blind. The purpose of this workshop is not as a venue for publication, so much as a place to gather together those in the community working on or interested in FGVC. Submissions of work which has been previously published, including papers accepted to the main CVPR 2017 conference are allowed (see below).
Submission of Previously Published Work:
We invite submissions of relevant work which has been previously published, including papers accepted to the main CVPR 2017 conference. The purpose of the workshop is not a venue for publication, so much as a place to gather together those in the community either working on or interested in fine-grained categorization. In the case of previously published work, it is not necessary for the authors to maintain anonymity. Instead, please cite the existing publication in the submitted abstract. These will be reviewed single-blind (much as a journal is reviewed: authors are known to reviewers, reviewers unknown to authors).