Latest News

  • 2020/04/17: We sincerely looking forward to paper submissions, HERE.
  • 2020/03/27: We appreciate that Rita Cucchiara, Weishi Zheng and Elisa Ricci have confirmed their keynote speeches.

Call for papers

The full-day workshop on Fine-Grained Visual Recognition and re-IDentification (FGVRID 2020) will be organized on January 11, 2021, in conjunction with ICPR 2020, The 25th International Conference on Pattern Recognition, Milan, Italy, January 10-15, 2021.

This workshop will bring together researchers from subfields of patter recognition that have seen growing activity in the past few years: fine-grained visual recognition and person/vehicle re-identification.

The ubiquitous surveillance cameras are generating huge amount of videos. Automatic video content analysis and recognition are thus desirable for effective utilization of those data. Fine-Grained Visual Recognition and Re-Identification (FGVRID) aims to accurately identify visual objects and match re-appearing targets, e.g., persons and vehicles from a large set of images and videos. It has the potential to offer an unprecedented possibility for intelligent video processing and analysis, as well as to explore the promising applications on public security.

The FGVRID workshop wishes to bring together researchers from fine-grained visual categorization, as well as person/ vehicle ReID communities, and to foster discussions and exchange of ideas between them. FGVRID is not a traditional search or classification task due to its goal of accurately identifying visual objects. First, proper detection algorithms should be designed to locate objects and their parts in videos before proceeding to the identification step. Second, the visual appearance of an object is easily affected by many factors like viewpoint changes and camera parameter differences, etc. Third, annotating the fine-grained identity or category cues is expensive and time consuming. Finally, to cope with the large-scale data, scalable indexing or feature coding algorithms should be designed to ensure the online recognition efficiency. Aiming to seek novel solutions and possibilities in FGVRID, this workshop will have in-depth discussions on those issues and aims to go beyond toy datasets and small-scale algorithms. Specifically, the covered topics include, but are not limited to:

  • Unsupervised, semi-supervised, and transfer learning algorithms
  • Robust object detection and tracking in the wild
  • Efficient and effective video representations
  • Object parsing and layout estimation
  • Large-scale indexing, feature coding, and retrieval algorithms
  • Fine-grained visual classification
  • New problems and datasets for fine-grained visual recognition and re-identification

Paper Formatting Instructions:
FGVRID Workshop follows a single-blind review process.  Authors are required to include their names and affiliations in their papers as illustrated in the sample templates.

Template for submission in Word or Latex can be downloaded here: Word and Latex Templates

Papers Submission:
Submissions should be made to the appropriate Conference tracks. All papers will be peer reviewed single-blind. The accepted papers will be published by IEEE and will be available in the IEEExplore.  Submissions should be 6 pages minimum, 8 pages maximum including references (IEEE format).

Submission Site for the FGVRID Workshop is available on MicroSoft CMT  HERE

Keynote speakers

Weishi Zheng

Weishi Zheng
Sun Yat-sen University
zhwshi@mail.sysu.edu.cn

Rita Cucchiara

Rita Cucchiara
University of Modena and Reggio Emilia
rita.cucchiara@unimore.it

Elisa Ricci

Elisa Ricci
University of Trento
e.ricci@unitn.it

TBD

Alan Hanjalic

Alan Hanjalic
Delft University of Technology
A.hanjalic@utdelft.nl

Cees G.M. Snoek

Cees G.M. Snoek
University of Amsterdam
cgmsnoek@uva.nl

Alberto Del Bimbo

Alberto Del Bimbo
University of Florence
alberto.delbimbo@unifi.it

Qi Tian

Qi Tian
Huawei Research
q-tian@hotmail.com

Accepted papers

TBD

Program

TBD

The workshop will include both invited talks and original submissions selected through a double-blind review process. The proposed length of this workshop is one day. The rough outline and highlights are as follows:

1. 3-5 speakers will be invited to give talks in the workshop. Each speaker will give a 45mins talk on fine-grained visual recognition and person/vehicle ReID or related topics.

2. Apart from the invited talks, there will be oral sessions and poster sessions. We plan to accept submissions not according to their achieved performance on public datasets, but rather encourage people to take on a research question related to the topic of the workshop. Examples of such questions are:

  • How can fine-grained visual recognition and person/vehicle re-identification benefit each other?
  • How much do object/part detections influence performance in fine-grained visual recognition/re-identification?
  • What are the most challenging issues in fine-grained visual recognition/re-identification?
  • How can fine-grained visual recognition/re-identification and Generative Adversarial Networks (GAN) benefit each other?
  • What are the limitations of existing public benchmark datasets?
  • How to tackle the large-scale fine-grained visual recognition/re-identification problem?

People involved

Organizers

Shiliang Zhang

Shiliang Zhang
Peking University
slzhang.jdl@pku.edu.cn

Guorong Li

Guorong Li
University of Chinese Academy of Sciences
liguorong@ucas.edu.cn

Weigang Zhang

Weigang Zhang
Harbin Institute of Technology, Weihai
wgzhang@hit.edu.cn

Qingming Huang

Qingming Huang
University of Chinese Academy of Sciences
qmhuang@ucas.ac.cn

Nicu Sebe

Nicu Sebe
University of Trento
niculae.sebe@unitn.it

Contact Us

For any information, please send an e-mail to Shiliang Zhang, Guorong Li and Weigang Zhang.


iapr    ieee