Call for Papers
Check it out!
Thank you very much for your participation!
Latest News
| June 23, 2015 | Photos of the workshop have been available. |
| June 23, 2015 | Information on award winner has been available. |
| October 9, 2014 | Presentation instruction have been available. |
| October 9, 2014 | Workshop program has been available. |
| September 28, 2014 | The list of accepted papers has been available. |
| September 23, 2014 | [Important] The author notification and camera-ready paper due have been extended. See important dates for more detail. |
| September 10, 2014 | [Important] The paper submission due has been extended to September 17. See important dates for more detail. |
| August 13, 2014 | Paper submission site has been open. |
| July 3, 2014 | [Important] The workshop is assigned to November 2. |
| June 3, 2014 | [Important] The important dates and Submission and Publication section has been updated. |
| May 15, 2014 | The website is temporarily open. |
Important Dates
| Paper Submission Due: | |
| Author Notification: | |
| Camera-Ready Paper Due: |
Workshop Overview
Background
Interpreting written communication (textual or symbolic) is one of the most important human activities. It is, however, one of the most difficult ones for computers to realize and has been recognized as an unsolved challenge. In the community of Document Analysis and Recognition (DAR), research on unconstrained reading systems has been of central interest. The series of ICDAR Robust Reading Competition (RRC), which is held in conjunction with biannual International Conference of Document Analysis and Recognition (ICDAR), has played an important role, especially, by providing standardized datasets that have become the de-facto evaluation benchmarks in the area and providing an opportunity to compare methods.
Recently, this kind of research, especially research on reading scene text, has attracted the interest and is increasingly accepted by the wider computer vision (CV) community. Papers on related topics have been consistently presented in major CV conferences such as CVPR, ICCV, ECCV and ACCV. In these papers, ICDAR RRC datasets are used as standard measures of methods.
Responding to this trend, we organize the First International Workshop on Robust Reading (IWRR 2014) in conjunction with ACCV 2014. The workshop aims at turning the spotlight on the challenge of unconstrained reading systems. The goal of the workshop is to attract the wider CV community to put research efforts on this challenging research topic, as well as to motivate many researchers from the DAR community to attend ACCV 2014.
Topics of Interest
include, but are not limited to:
- Scene text localisation, segmentation, and recognition
- Reading scene and/or overlaid text in video sequences
- Text localisation, segmentation, and recognition in born-digital images
- Restoration of camera captured documents (dewarping, deblurring, etc.)
- Quality estimation and degradation modelling of camera-captured text
- Performance evaluation and metrics
- Applications such as translation, reading text for the blind, etc
- Graphical content interpretation in complex settings
Keynote Talks
Text Reading in the Wild - How to make it useful?
- Lukáš Neumann (Czech Technical University in Prague, Czech Republic)
- Lukas Neumann is a PhD student at the the Center for Machine Perception of Czech Technical University in Prague, Czech Republic. Lukas received the MSc degree in system programming (with honors) from the Czech Technical University in 2010. His main research topic is unconstrained text localization and recognition in real-world images and video. Lukas has published at major computer vision conferences (ICCV, CVPR, ECCV, ACCV) and he received the best student paper price at the 2013 International Conference on Document Analysis and Recognition (ICDAR). In 2013 Lukas received a highly prestigious Google PhD Fellowship in computer vision.
- Prof. Jiří Matas (Czech Technical University in Prague, Czech Republic)
- Jiri Matas is Professor at the the Center for Machine Perception of Czech Technical University in Prague, Czech Republic. Jiri received the MSc degree in cybernetics (with honors) from the Czech Technical University in 1987 and the PhD degree from the University of Surrey, UK, in 1995. His research interests include object recognition, image retrieval, tracking, sequential pattern recognition, invariant feature detection, and Hough Transform and RANSAC-type optimization. Jiri has published more than 150 papers in refereed journals and conferences. His publications have approximately 17,000 citations in Google scholar, and his h-index is 48. He received the best paper prize at the British Machine Vision Conferences in 2002 and 2005 and at the Asian Conference on Computer Vision in 2007. Jiri Matas has served in various roles at major international conferences (e.g. ICCV, CVPR, ICPR, NIPS, ECCV), co-chairing ECCV 2004 and CVPR 2007. He is on the editorial board of IJCV.
PhotoOCR: Reading Text in Uncontrolled Conditions
- Dr. Alessandro Bissacco (Google Inc.)
- Alessandro Bissacco is a software engineer at Google since January 2007. He has worked on projects involving image matching, landmark recognition, object detection, text detection and OCR. His contributions are in use in several Google services such as Streetview, Google Goggles and Image Search. Currently he leads Google efforts on developing new technology for reading text from camera images in unconstrained environments, such as Google Goggles and Streetview. He holds a PhD in Computer Science from University of California, Los Angeles.
Submission and Publication
Submission to IWRR
The workshop organizers welcome original technical papers related to the workshop topics. All submissions must follow the guideline of ACCV 2014. Namely, submissions must be a maximum of 14 pages (not including references). Supplementary material can be submitted. Submissions, including supplementary material, must be self-contained. Submitted papers will be reviewed by the workshop program committee in a double blind manner (so all submissions must be anonymised). Papers violating the formatting rules will be rejected without review.
Go to Paper Submission Site (CMT)
Submission to ACCV2014 and designate IWRR as the "optional workshop"
ACCV2014 takes the "fall back" strategy, which provides opportunities for papers rejected from the main conference to be accepted by appropriate workshops. The papers rejected by the main conference but selected a workshop by the author(s) will be automatically passed to the workshop organizers, together with the main conference review comments. The workshop organizers will decide whether the papers can be accepted or not by the workshop, based on the paper and the reviews.
For the purpose, please select IWRR as the optional workshop, when submitting a related paper to the IWRR topic to the main conference.
Publication
Accepted papers will be included in the workshop proceedings, which will be published by Springer in the Lecture Notes in Computer Science (LNCS) series and also contained in the USB stick distributed in the conference site. Accepted papers must be registered and presented; otherwise they will not be included in the workshop proceedings.
Presentation Instruction
Oral Presentation
Each oral presentation has 20 min. including QA. Please check the length of your presentation and consider to finish your presentation 3-5 minutes advance in order to accept questions.
Best Paper Award - Sponsored by Xerox
One award of S$ 250 will be given to the best paper of the workshop, to be announced at the end of the workshop.
Presentations are judged in terms of
- Novelty
- Potential impact
- Presentation quality
Award Winner
The award is given to
Yu Zhou, Shuang Liu, Yongzheng Zhang, Yipeng Wang (Chinese Academy of Sciences) and Weiyao Lin (Shanghai Jiao Tong University)
who presented the paper entitled
Perspective Scene Text Recognition with Feature Compression and Ranking.
Congratulations!
